Benchmarking Methodology for Network Security Device Performance

Introduction 18 years have passed since IETF initially recommended test methodology and terminology for firewalls (). Firewalls have evolved significantly from the days of simple ACL filters. As the underlying technology progresses and improves, recommending test methodology and terminology for firewalls, requirements, and expectations for network security elements has increased tremendously. Security function implementations have evolved and diversified into intrusion detection and prevention, threat management, analysis of encrypted traffic, and more. In an industry of growing importance, well-defined and reproducible key performance indicators (KPIs) are increasingly needed to enable fair and reasonable comparison of network security functions. These reasons led to the creation of a new next-generation network security device benchmarking document, which makes obsolete. Measurement of performance for processing of IP fragmented traffic (see Section 5.9 of ) was not included in this document since IP fragmentation does today not commonly occur in traffic anymore, unlike it might have been at the time when was written. It should also be noted that retains significant value and has been consulted frequently while creating this document. For a more detailed explanation of what an NGFW is see the Wikipedia article .

Requirements The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 , when, and only when, they appear in all capitals, as shown here.

Scope This document provides testing terminology and testing methodology for modern and next-generation network security devices that are configured in Active ("Inline", see and ) mode. It covers the validation of security effectiveness configurations of network security devices, followed by performance benchmark testing. This document focuses on advanced, realistic, and reproducible testing methods. Additionally, it describes testbed environments, test tool requirements, and test result formats.

Test Setup The test setup defined in this document applies to all benchmarking tests described in . The test setup MUST be contained within an Isolated Test Environment (see Section 3 of ).

Testbed Configuration Testbed configuration MUST ensure that any performance implications that are discovered during the benchmark testing aren't due to the inherent physical network limitations such as the number of physical links and forwarding performance capabilities (throughput and latency) of the network devices in the testbed. For this reason, this document recommends avoiding external devices such as switches and routers in the testbed wherever possible. In some deployment scenarios, the network security devices (Device Under Test/System Under Test) are connected to routers and switches, which will reduce the number of entries in MAC or ARP/ND (Address Resolution Protocol/ Neighbor Discovery) tables of the Device Under Test/System Under Test (DUT/SUT). If MAC or ARP/ND tables have many entries, this may impact the actual DUT/SUT performance due to MAC and ARP/ND table lookup processes. This document also recommends using test equipment with the capability of emulating layer 3 routing functionality instead of adding external routers in the testbed. The testbed setup Option 1 () is the RECOMMENDED testbed setup for the benchmarking test.

Testbed Setup - Option 1 If the test equipment used is not capable of emulating OSI layer 3 routing functionality or if the number of used ports is mismatched between test equipment and the DUT/SUT (need for test equipment port aggregation), the test setup can be configured as shown in .

Testbed Setup - Option 2

DUT/SUT Configuration The same DUT/SUT configuration MUST be used for all benchmarking tests described in . Since each DUT/SUT will have its own unique configuration, users MUST configure their devices with the same parameters and security features that would be used in the actual deployment of the device or a typical deployment. The DUT/SUT MUST be configured in "Inline" mode so that the traffic is actively inspected by the DUT/SUT. and below describe the RECOMMENDED and OPTIONAL sets of network security features for NGFW and NGIPS, respectively. The selected security features MUST be consistently enabled on the DUT/SUT for all benchmarking tests described in . To improve repeatability, a summary of the DUT/SUT configuration including a description of all enabled DUT/SUT features MUST be published with the benchmarking results. The following table provides a brief description of the security features and these are approximate taxonomies of features commonly found in currently deployed NGFW and NGIDS. The features provided by specific implementations may be named differently and not necessarily have configuration settings that align with the taxonomy. Security Feature Description

DUT/SUT Features	Description
TLS Inspection	DUT/SUT intercepts and decrypts inbound HTTPS traffic between servers and clients. Once the content inspection has been completed, DUT/SUT encrypts the HTTPS traffic with ciphers and keys used by the clients and servers. For TLS1.3, the DUT works as a middlebox (proxy) and it holds the certificates and Pre-Shared Keys (PSK) that are trusted by the client and represent the identity of the real server.
IDS/IPS	DUT/SUT detects and blocks exploits targeting known and unknown vulnerabilities across the monitored network.
Anti-Malware	DUT/SUT detects and prevents the transmission of malicious executable code and any associated communications across the monitored network. This includes data exfiltration as well as command and control channels.
Anti-Spyware	Anti-Spyware is a subcategory of Anti Malware. Spyware transmits information without the user's knowledge or permission. DUT/SUT detects and blocks initial infection or transmission of data.
Anti-Botnet	DUT/SUT detects and blocks traffic to or from botnets.
Anti-Evasion	DUT/SUT detects and mitigates attacks that have been obfuscated in some manner.
Web Filtering	DUT/SUT detects and blocks malicious websites including defined classifications of websites across the monitored network.
DLP	DUT/SUT detects and prevents data breaches and data exfiltration, or it detects and blocks the transmission of sensitive data across the monitored network.
Certificate Validation	DUT/SUT validates certificates used in encrypted communications across the monitored network.
Logging and Reporting	DUT/SUT logs and reports all traffic at the flow level across the monitored network.
Application Identification	DUT/SUT detects known applications as defined within the traffic mix selected across the monitored network.
DPI	DUT/SUT inspects the content of the data packet.

NGFW Security Features

DUT/SUT (NGFW) Features	RECOMMENDED	OPTIONAL
TLS Inspection	x
IDS/IPS	x
Anti-Spyware	x
Anti-Virus	x
Anti-Botnet	x
Web Filtering		x
Data Loss Protection (DLP)		x
DDoS Protection		x
Certificate Validation		x
Logging and Reporting	x
Application Identification	x

NGIPS Security Features

DUT/SUT (NGIPS) Features	RECOMMENDED	OPTIONAL
TLS Inspection	x
Anti-Malware	x
Anti-Spyware	x
Anti-Botnet	x
Application Identification	x
Deep Packet Inspection (DPI)	x
Anti-Evasion	x

Note: With respect to TLS Inspection, there are scenarios where it will be optional. Below is a summary of the DUT/SUT configuration:

DUT/SUT MUST be configured in "inline" mode.
"Fail-Open" behavior MUST be disabled.
All RECOMMENDED security features are enabled.
Logging and reporting MUST be enabled. DUT/SUT SHOULD log all traffic at the flow level - Logging to an external device is permissible.
Geographical location filtering SHOULD be configured.
Application Identification and Control MUST be configured to trigger application from the defined traffic mix.

In addition, a realistic number of access control rules (ACL) SHOULD be configured on the DUT/SUT where ACLs are configurable and reasonable based on the deployment scenario. For example, it is acceptable not to configure ACLs in an NGIPS since NGIPS devices do not require the use of ACLs in most deployment scenarios. This document determines the number of access policy rules for four different classes of DUT/SUT: Extra Small (XS), Small (S), Medium (M), and Large (L). A sample DUT/SUT classification is described in . The Access Control Rules (ACL) defined in MUST be configured from top to bottom in the correct order as shown in the table. This is due to ACL types listed in specificity decreasing order, with "block" first, followed by "allow", representing a typical ACL-based security policy. The ACL entries MUST be configured with routable IP prefixes by the DUT/SUT, where applicable. (Note: There will be differences between how security vendors implement ACL decision-making.) The configured ACL MUST NOT block the test traffic used for the benchmarking tests.

Security Effectiveness Configuration The selected security features (defined in and ) of the DUT/SUT MUST be configured effectively to detect, prevent, and report the defined security vulnerability sets. This section defines the selection of the security vulnerability sets from the Common Vulnerabilities and Exposures (CVE) list for testing. The vulnerability set should reflect a minimum of 500 CVEs from no older than 10 calendar years to the current year. These CVEs should be selected with a focus on in-use software commonly found in business applications, with a Common Vulnerability Scoring System (CVSS) Severity of High (7-10). This document is primarily focused on performance benchmarking. However, it is RECOMMENDED to validate the security features configuration of the DUT/SUT by evaluating the security effectiveness as a prerequisite for performance benchmarking tests defined in section 7. In case the benchmarking tests are performed without evaluating security effectiveness, the test report MUST explain the implications of this. The methodology for evaluating security effectiveness is defined in .

Test Equipment Configuration In general, test equipment allows configuring parameters in different protocol layers. Extensive proof of concept tests conducted to support preparation of this document showed that benchmarking results are strongly affected by the choice of protocol stack parameters; especially OSI layer 4 transport protocol parameters. For more information on how TCP and QUIC parameters will impact performance review . To achieve reproducible results that will be representative for real deployment scenarios, careful specification and documentation of the parameters are required. This section specifies common test equipment configuration parameters applicable for all benchmarking tests defined in . Any benchmarking test specific parameters are described under the test setup section of each benchmarking test individually.

Client Configuration This section specifies which parameters should be considered while configuring emulated client endpoints in the test equipment. Also, this section specifies the RECOMMENDED values for certain parameters. The values are the defaults typically used in most of the client operating system types. Pre-standard evaluations have shown that it is possible to set a wide range of arbitrary parameters for OSI layer 4 transport protocols on test equipment leading to client-specific results optimization; however, only well-defined common parameter sets help to establish meaningful and comparable benchmarking results. For these reasons, this document recommends specific sets of transport protocol parameters to be configured on test equipment used for benchmarking.

TCP Stack Attributes The TCP stack of the emulated client endpoints MUST fulfill the TCP requirements defined in (See Appendix B.). In addition, this section specifies the RECOMMENDED values for TCP parameters configured using the following parameters: The IPv4 and IPv6 Maximum Segment Size (MSS) are set to 1460 bytes and 1440 bytes respectively. TX and RX initial receive window sizes are set to 65535 bytes. The client's initial congestion window should not exceed 10 times the MSS. Delayed ACKs are permitted and the maximum client delayed ACK should not exceed 10 times of the MSS before a forced ACK also, the maximum delayed ACK timer is allowed to be set to 200 ms. Up to three retries are allowed before a timeout event is declared. TCP PSH flag is set to high in all traffic. The source port range is in the range of 1024 - 65535. The clients initiate TCP connections via a three-way handshake (SYN, SYN/ACK, ACK) and close TCP connections via either a TCP three-way close (FIN, FIN/ACK, ACK) or a TCP four-way close (FIN, ACK, FIN, ACK).

QUIC Specification QUIC stack emulation on the test equipment MUST conform to and . This section specifies the RECOMMENDED values for certain QUIC parameters to be configured on test equipment used for benchmarking purposes only. QUIC Stream type (defined in section 2.1 of ) is set to "Client-Initiated, Bidirectional". 0-RTT and early data are Disabled. QUIC Connection termination method is Immediate close (section 10.2 of . Flow control is enabled. UDP payloads are set to datagram size of 1232 bytes for IPv6 and 1252 bytes for IPv4. In addition, transport parameters and default values defined in section 18.2 of are RECOMMENDED to configure on test equipment. Also, this document references Appendixes B.1 and B.2 of for congestion control related constants and variables. Any configured QUIC and UDP parameter(s) MUST be documented in the test report.

Client IP Address Space The client IP space contains the following attributes.

If multiple IP blocks are used, they MUST be consist of multiple unique, discontinuous static address blocks.
A default gateway MAY be used.
The DSCP (differentiated services code point) marking should be set to DF (Default Forwarding) '000000' on IPv4 Type of Service (ToS) field and IPv6 traffic class field.
Extension header(s) MAY be used for IPv6 clients. If multiple extension headers are needed for traffic emulation, this document references to choose the correct order of the extension headers within an IPv6 packet. Testing with extension header(s) may impact the performance of the DUT. The extension headers MUST be documented and reported.

The following equation can be used to define the total number of client IP addresses that need to be configured on the test equipment. Desired total number of client IP addresses = Target throughput [Mbit/s] / Average throughput per IP address [Mbit/s] As shown in the example list below, the value for "Average throughput per IP address" can be varied depending on the deployment and use case scenario.

DUT/SUT deployment scenario 1 : 6-7 Mbit/s per IP (e.g. 1,400-1,700 IPs per 10Gbit/s of throughput)
DUT/SUT deployment scenario 2 : 0.1-0.2 Mbit/s per IP (e.g. 50,000-100,000 IPs per 10Gbit/s of throughput)

Based on deployment and use case scenario, client IP addresses SHOULD be distributed between IPv4 and IPv6. The following options MAY be considered for a selection of ratios for both IP addresses and traffic load distribution.

100 % IPv4, no IPv6
80 % IPv4, 20% IPv6
50 % IPv4, 50% IPv6
20 % IPv4, 80% IPv6
no IPv4, 100% IPv6

Note: IANA has assigned IP address ranges for testing purposes as described in . If the test scenario requires more IP addresses or subnets than IANA has assigned, this document recommends using private IPv4 address ranges or Unique Local Address (ULA) IPv6 address ranges for the testing.

Emulated Web Browser Attributes The client emulated web browser (emulated browser) contains attributes that will materially affect the traffic load. The objective is to emulate modern, typical browser attributes to improve the relevance of the result set for typical deployment scenarios. The emulated browser MUST negotiate HTTP version 1.1 or higher. The emulated browser SHOULD advertise a User-Agent header. The emulated browser MUST enforce content length validation. HTTP header compression MAY be set to enable. If HTTP header compression is configurable in the test equipment, it MUST be documented if it was enabled or disabled. Depending on test scenarios and chosen HTTP version, the emulated browser MAY open multiple TCP or QUIC connections per Server endpoint IP at any time depending on how many sequential transactions need to be processed. For HTTP/2 traffic emulation, the emulated browser opens multiple concurrent streams per connection (multiplexing). For HTTPS requests, the emulated browser MUST send "h2" protocol identifier using the TLS extension Application Layer Protocol Negotiation (ALPN). The following default values (see ) are the RECOMMENDED setting for certain HTTP/2 parameters to be configured on test equipment used for benchmarking purposes only:

Maximum Frame size: 16384 bytes
Initial Window size: 65535 bytes
HPACK Header table size: 4096 bytes
Server PUSH enable: false (Note: in the default setting is true. However, for testing purposes, this document recommends setting the value false for server push.)

This document refers to for further details of HTTP/2. If any additional parameters are used to configure the test equipment, it MUST be documented. For HTTP/3 traffic emulation, the emulated browsers initiate secure QUIC connections using TLS 1.3 ( describes how TLS is used to secure QUIC). This document refers to for HTTP/3 specifications. The specification for transport protocol parameters is defined in . QPACK configuration settings such as MAX_TABLE_CAPACITY and QPACK_BLOCKED_STREAMS are set to zero (default) as defined in . Any HTTP/3 parameters used for test equipment configuration MUST be documented. For encrypted traffic, the following attributes are defined as the negotiated encryption parameters. The test clients MUST use TLS version 1.2 or higher. The TLS record size MAY be optimized for the HTTPS response object size up to a record size of 16 KBytes. If Server Name Indication (SNI) is required (especially if the server is identified by a domain name), the client endpoint MUST send TLS extension Server Name Indication (SNI) information when opening a security tunnel. Each client connection MUST perform a full TLS handshake and session reuse or resumption MUST be disabled. (Note: Real web browsers use session reuse or resumption. However, for testing purposes, this feature must not be used to measure the DUT/SUT performance in the worst-case scenario.) The following TLS 1.2 supported ciphers and keys are RECOMMENDED for HTTPS based benchmarking tests defined in .

ECDHE-ECDSA-AES128-GCM-SHA256 with Prime256v1 (Signature Hash Algorithm: ecdsa_secp256r1_sha256 and Supported group: secp256r1)
ECDHE-RSA-AES128-GCM-SHA256 with RSA 2048 (Signature Hash Algorithm: rsa_pkcs1_sha256 and Supported group: secp256r1)
ECDHE-ECDSA-AES256-GCM-SHA384 with Secp384r1 (Signature Hash Algorithm: ecdsa_secp384r1_sha384 and Supported group: secp384r1)
ECDHE-RSA-AES256-GCM-SHA384 with RSA 4096 (Signature Hash Algorithm: rsa_pkcs1_sha384 and Supported group: secp384r1)

Note: The above ciphers and keys were those commonly used for enterprise-grade encryption cipher suites for TLS 1.2 as of the time of publication (2022). Individual certification bodies should use ciphers and keys that reflect evolving use cases. These choices MUST be documented in the resulting test reports with detailed information on the ciphers and keys used along with reasons for the choices. IANA recommends the following cipher suites for use with TLS 1.3 defined in .

TLS_AES_128_GCM_SHA256
TLS_AES_256_GCM_SHA384
TLS_CHACHA20_POLY1305_SHA256
TLS_AES_128_CCM_SHA256

Backend Server Configuration This section specifies which parameters should be considered while configuring emulated backend servers using test equipment.

TCP Stack Attributes The TCP stack on the server-side MUST be configured similar to the client-side configuration described in

QUIC Specification The QUIC parameters on the server-side MUST be configured similar to the client-side configuration. Any configured QUIC Parameter(s) MUST be documented in the report.

Server Endpoint IP Addressing The sum of the server IP space MUST contain the following attributes.

The server IP blocks MUST consist of unique, discontinuous static address blocks with one IP per server Fully Qualified Domain Name (FQDN) endpoint per test port.
A default gateway is permitted. The DSCP (differentiated services code point) marking is set to DF (Default Forwarding) '000000' on IPv4 Type of Service (ToS) field and IPv6 traffic class field. Extension header(s) for the IPv6 server is permitted. If multiple extension headers are required, this document referenced to choose the correct order of the extension headers within an IPv6 packet.
The server IP address distribution between IPv4 and IPv6 MUST be identical to the client IP address distribution ratio.

Note: The IANA has assigned IP address blocks for the testing purpose as described in . If the test scenario requires more IP addresses or address blocks than the IANA assigned, this document recommends using private IPv4 address ranges or Unique Local Address (ULA) IPv6 address ranges for the testing.

HTTP / HTTPS Server Pool Endpoint Attributes The HTTP 1.1 and HTTP/2 server pools listen on TCP ports 80 and 443 for HTTP and HTTPS. HTTP/3 server pool listens on UDP port 443 or any port. The server MUST emulate the same HTTP version (HTTP 1.1 or HTTP/2 or HTTP/3) and settings chosen by the client (emulated web browser). For the HTTPS server, TLS 1.2 or higher MUST be used with a maximum record size of 16 KByte. Ticket resumption or session ID reuse MUST NOT be used for TLS 1.2 and also session Ticket or session cache MUST NOT be used for TLS 1.3. The server MUST serve a certificate to the client. Cipher suite and key size on the server-side MUST be configured similar to the client-side configuration described in .

Traffic Flow Definition This section describes the traffic pattern between client and server endpoints. At the beginning of the test, the server endpoint initializes and will be ready to accept connection states including initialization of the TCP or QUIC stack as well as bound HTTP and HTTPS servers. When a client endpoint is needed, it will initialize and be given attributes such as a MAC and IP address. The behavior of the client is to sweep through the given server IP space, generating a recognizable service by the DUT. Sequential and pseudorandom sweep methods are acceptable. The method used MUST be stated in the final report. Thus, a balanced mesh between client endpoints and server endpoints will be generated in a client IP and port to server IP and port combination. Each client endpoint performs the same actions as other endpoints, with the difference being the source IP of the client endpoint and the target server IP pool. The client MUST use the server IP address or FQDN in the host header.

Description of Intra-Client Behavior Client endpoints are independent of other clients that are concurrently executing. When a client endpoint initiates traffic, this section describes how the client steps through different services. Once the test is initialized, the client endpoints randomly hold (perform no operation) for a few milliseconds for better randomization of the start of client traffic. Each client (HTTP 1.1 or HTTP/2) will either open a new TCP connection or connect to an HTTP persistent connection still open to that specific server. HTTP/3 clients will open UDP streams within QUIC connections. At any point that the traffic profile may require encryption, a TLS encryption tunnel will form presenting the URL or IP address request to the server. If using SNI, the server MUST then perform an SNI name check with the proposed FQDN compared to the domain embedded in the certificate. Only when correct, will the server process the HTTPS response object. The initial response object to the server is based on benchmarking tests described in . Multiple additional sub-URLs (response objects on the service page) MAY be requested simultaneously. This MAY be to the same server IP as the initial URL. Each sub-object will also use a canonical FQDN and URL path.

Traffic Load Profile The loading of traffic is described in this section. The loading of a traffic load profile has five phases: Init, ramp up, sustain, ramp down, and collection.

Init phase: Testbed devices including the client and server endpoints should negotiate layer 2-3 connectivity such as MAC learning and ARP/ND. Only after successful MAC learning or ARP/ND SHALL the test iteration move to the next phase. No measurements are made in this phase. The minimum RECOMMENDED time for the Init phase is 5 seconds. During this phase, the emulated clients MUST NOT initiate any sessions with the DUT/SUT, in contrast, the emulated servers should be ready to accept requests from DUT/SUT or emulated clients.
Ramp up phase: The test equipment MUST start to generate the test traffic. It MUST use a set of the approximate number of unique client IP addresses to generate traffic. The traffic MUST ramp up from zero to desired target objective. The target objective is defined for each benchmarking test. The duration for the ramp up phase MUST be configured long enough that the test equipment does not overwhelm the DUT/SUTs stated performance metrics defined in namely, TCP or QUIC Connections Per Second, Inspected Throughput, Concurrent TCP or QUIC Connections, and Application Transactions Per Second. No measurements are made in this phase.
Sustain phase: Starts when all required clients are active and operating at their desired load condition. In the sustain phase, the test equipment MUST continue generating traffic to a constant target value for a constant number of active clients. The minimum RECOMMENDED time duration for sustain phase is 300 seconds. This is the phase where measurements occur. The test equipment MUST measure and record statistics continuously. The sampling interval for collecting the raw results and calculating the statistics MUST be less than 2 seconds.
Ramp down phase: The test traffic slows down from the target number to 0, and no measurements are made. The time duration for ramp up and ramp down phases MUST be the same.
Collection phase: The last phase is administrative and will occur when the test equipment merges and collates the report data.

Testbed Considerations This section describes steps for a reference test (pre-test) that control the test environment including test equipment, focusing on physical and virtualized environments and as well as test equipment. Below are the RECOMMENDED steps for the reference test.

Perform the reference test either by configuring the DUT/SUT in the most trivial setup (fast forwarding) or without the presence of the DUT/SUT.
Generate traffic from traffic generator. Choose a traffic profile used for the HTTP or HTTPS throughput performance test with the smallest object size.
Ensure that any ancillary switching or routing functions added in the test equipment do not limit the performance by introducing network metrics such as packet loss and latency. This is specifically important for virtualized components (e.g., vSwitches, vRouters).
Verify that the generated traffic (performance) of the test equipment matches and reasonably exceeds the expected maximum performance of the DUT/SUT.
Record the network performance metrics packet loss and latency introduced by the test environment (without DUT/SUT).
Assert that the testbed characteristics are stable during the entire test session. Several factors might influence stability specifically, for virtualized testbeds. For example, additional workloads in a virtualized system, load balancing, and movement of virtual machines during the test, or simple issues such as additional heat created by high workloads leading to an emergency CPU performance reduction.

The reference test MUST be performed before the benchmarking tests (described in section 7) start.

Reporting This section describes how the benchmarking test report should be formatted and presented. It is RECOMMENDED to include two main sections in the report: the introduction and the detailed test results sections.

Introduction The following attributes should be present in the introduction section of the test report.

The time and date of the execution of the tests
Summary of testbed software and hardware details
1. DUT/SUT hardware/virtual configuration
  - This section should clearly identify the make and model of the DUT/SUT
  - The port interfaces, including speed and link information
  - If the DUT/SUT is a Virtual Network Function (VNF), host (server) hardware and software details, interface acceleration type such as DPDK and SR-IOV, used CPU cores, used RAM, resource sharing (e.g. Pinning details and NUMA Node) configuration details, hypervisor version, virtual switch version
  - details of any additional hardware relevant to the DUT/SUT such as controllers
2. DUT/SUT software
  - Operating system name
  - Version
  - Specific configuration details (if any)
3. DUT/SUT enabled features
  - Configured DUT/SUT features (see and )
  - Attributes of the above-mentioned features
  - Any additional relevant information about the features
4. Test equipment hardware and software
  - Test equipment vendor name
  - Hardware details including model number, interface type
  - Test equipment firmware and test application software version
  - If the test equipment is a virtual solution, the host (server) hardware and software details, interface acceleration type such as DPDK and SR-IOV, used CPU cores, used RAM, resource sharing (e.g. Pinning details and NUMA Node) configuration details, hypervisor version, virtual switch version
5. Key test parameters
  - Used cipher suites and keys
  - IPv4 and IPv6 traffic distribution
  - Number of configured ACL
  - TCP, UDP stack parameter if tested
  - QUIC, HTTP/2, and HTTP/3 parameters if tested
6. Details of application traffic mix used in the benchmarking test "Throughput Performance with Application Traffic Mix"
  - Name of applications and layer 7 protocols
  - Percentage of emulated traffic for each application and layer 7 protocols
  - Percentage of encrypted traffic and used cipher suites and keys (The RECOMMENDED ciphers and keys are defined in )
  - Used object sizes for each application and layer 7 protocols
Results Summary / Executive Summary
1. Results should be presented with an introduction section documenting the summary of results in a prominent, easy to read block.

Detailed Test Results In the result section of the test report, the following attributes should be present for each benchmarking test.

KPIs MUST be documented separately for each benchmarking test. The format of the KPI metrics MUST be presented as described in .
The next level of detail should be graphs showing each of these metrics over the duration (sustain phase) of the test. This allows the user to see the measured performance stability changes over time.

Benchmarks and Key Performance Indicators This section lists key performance indicators (KPIs) for overall benchmarking tests. All KPIs MUST be measured during the sustain phase of the traffic load profile described in . Also, the KPIs MUST be measured from the result output of test equipment.

Concurrent TCP Connections The aggregate number of simultaneous connections between hosts across the DUT/SUT, or between hosts and the DUT/SUT (defined in ).
Concurrent QUIC Connections The aggregate number of simultaneous connections between hosts across the DUT/SUT.
TCP Connections Per Second The average number of successfully established TCP connections per second between hosts across the DUT/SUT, or between hosts and the DUT/SUT. As described in , the TCP connections are initiated by clients via a TCP three-way handshake (SYN, SYN/ACK, ACK). Then the TCP session data is sent and then the TCP sessions are closed via either a TCP three-way close (FIN, FIN/ACK, ACK) or a TCP four-way close (FIN, ACK, FIN, ACK). The TCP sessions MUST NOT be closed by RST.
QUIC Connections Per Second The average number of successfully established QUIC connections per second between hosts across the DUT/SUT. As described in , the QUIC connections are initiated by clients. Then the data is sent and then the QUIC sessions are closed by "immediate close" method. Since QUIC specification defined in recommends disabling 0-RTT and early data, this KPI focused on 1-RTT handshake. If required, 0-RTT can be also measured in separate test runs while enabling 0-RTT and early data in the test equipment.
Application Transactions Per Second The average number of successfully completed transactions per second. For a particular transaction to be considered successful, all data MUST have been transferred in its entirety. In case of HTTP(S) transactions, it MUST have a valid status code (200 OK).
TLS Handshake Rate The average number of successfully established TLS connections per second between hosts across the DUT/SUT, or between hosts and the DUT/SUT. For TLS1.3 the handshake rate can be measured with 0-RTT or 1-RTT handshake. The transport protocol can be either TCP or QUIC.
Inspected Throughput The number of bits per second of examined and allowed traffic a network security device is able to transmit to the correct destination interface(s) in response to a specified offered load. The throughput benchmarking tests defined in SHOULD measure the average Layer 2 throughput value when the DUT/SUT is "inspecting" traffic. This document recommends presenting the inspected throughput value in Gbit/s rounded to two places of precision with a more specific Kbit/s in parenthesis.
Time to First Byte (TTFB) TTFB is the elapsed time between the start of sending the TCP SYN packet or QUIC initial Client Hello from the client and the client receiving the first packet of application data from the server via DUT/SUT. The benchmarking tests HTTP Transaction Latency and HTTPS Transaction Latency measure the minimum, average and maximum TTFB. The value should be expressed in milliseconds.
URL Response time / Time to Last Byte (TTLB) URL Response time / TTLB is the elapsed time between the start of sending the TCP SYN packet or QUIC initial Client Hello from the client and the client receiving the last packet of application data from the server via DUT/SUT. The benchmarking tests HTTP Transaction Latency and HTTPS Transaction Latency measure the minimum, average and maximum TTLB. The value should be expressed in milliseconds.

Benchmarking Tests This section mainly focuses on the benchmarking tests with HTTP/1.1 or HTTP/2 traffic which uses TCP as the transport protocol. In particular, this section does not define specific benchmarking tests for QUIC or HTTP/3 related KPIs. However, the test methodology defined in the benchmarking tests TCP/QUIC Connections Per Second with HTTPS Traffic, HTTPS Transaction Latency, HTTPS Throughput, and Concurrent TCP/QUIC Connection Capacity with HTTPS Traffic can be used to test QUIC or HTTP/3 related KPIs. The throughput performance test with application traffic mix defined in can be performed with any other application traffic including HTTP/3.

Throughput Performance with Application Traffic Mix

Objective Using a relevant application traffic mix, determine the sustainable inspected throughput supported by the DUT/SUT. Based on the test customer's specific use case, testers can choose the relevant application traffic mix for this test. The details about the traffic mix MUST be documented in the report. At least the following traffic mix details MUST be documented and reported together with the test results:

Name of applications and layer 7 protocols
Percentage of emulated traffic for each application and layer 7 protocol
Percentage of encrypted traffic and used cipher suites and keys (The RECOMMENDED ciphers and keys are defined in .)
Used object sizes for each application and layer 7 protocols

Test Setup Testbed setup MUST be configured as defined in . Any benchmarking test specific testbed configuration changes MUST be documented.

Test Parameters In this section, the benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented. In case the DUT/SUT is configured without TLS inspection, the test report MUST explain the implications of this to the relevant application traffic mix encrypted traffic.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test:

Client IP address ranges defined in
Server IP address ranges defined in
Traffic distribution ratio between IPv4 and IPv6 defined in
Target inspected throughput: Aggregated line rate of the interface(s) used in the DUT/SUT or the value defined based on the requirement for a specific deployment scenario
Initial throughput: 10% of the "Target inspected throughput" Note: Initial throughput is not a KPI to report. This value is configured on the traffic generator and used to perform Step 1: "Test Initialization and Qualification" described under .
One of the ciphers and keys defined in are RECOMMENDED to use for this benchmarking test.

Traffic Profile Traffic profile: This test MUST be run with a relevant application traffic mix profile.

Test Results Validation Criteria The following criteria are the test results validation criteria. The test results validation criteria MUST be monitored during the whole sustain phase of the traffic load profile.

Number of failed application transactions MUST be less than 0.001% (1 out of 100,000 transactions) of total attempted transactions.
Number of Terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections.
If HTTP/3 is used, the number of failed QUIC connections due to unexpected HTTP/3 error codes MUST be less than 0.001% (1 out of 100,000 connections) of total initiated QUIC connections.

Measurement Following KPI metrics MUST be reported for this benchmarking test: Mandatory KPIs (benchmarks): Inspected Throughput and Application Transactions Per Second Note: TTLB MUST be reported along with the object size used in the traffic profile. Optional TCP stack related KPIs: TCP Connections Per Second, TLS Handshake Rate, TTFB (minimum, average, and maximum), TTLB (minimum, average, and maximum) Optional QUIC stack related KPIs: QUIC connection per second and concurrent QUIC connections

Test Procedures and Expected Results The test procedures are designed to measure the inspected throughput performance of the DUT/SUT at the sustaining period of the traffic load profile. The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (initial throughput) and meets the test results validation criteria when it was very minimally utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IP types: IPv4 only, IPv6 only, and IPv4 and IPv6 mixed traffic distribution.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to generate test traffic at the "Initial throughput" rate as described in . The test equipment MUST follow the traffic load profile definition as described in . The DUT/SUT MUST reach the "Initial throughput" during the sustain phase. Measure all KPI as defined in . The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to step 2.

Step 2: Test Run with Target Objective Configure test equipment to generate traffic at the "Target inspected throughput" rate defined in . The test equipment MUST follow the traffic load profile definition as described in . The test equipment MUST start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective ("Target inspected throughput") in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable average inspected throughput within the test results validation criteria. The final test iteration MUST be performed for the test duration defined in .

TCP/HTTP Connections Per Second

Objective Using HTTP traffic, determine the sustainable TCP connection establishment rate supported by the DUT/SUT under different throughput load conditions. To measure connections per second, test iterations MUST use different fixed HTTP response object sizes (the different load conditions) defined in .

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in Target connections per second: Initial value from product datasheet or the value defined based on the requirement for a specific deployment scenario Initial connections per second: 10% of "Target connections per second" (Note: Initial connections per second is not a KPI to report. This value is configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under .) The client MUST negotiate HTTP and close the connection with FIN immediately after the completion of one transaction. In each test iteration, the client MUST send a GET request requesting a fixed HTTP response object size. The RECOMMENDED response object sizes are 1, 2, 4, 16, and 64 KByte.

Test Results Validation Criteria The following criteria are the test results validation criteria. The Test results validation criteria MUST be monitored during the whole sustain phase of the traffic load profile.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of total attempted transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent TCP connections MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. This confirms the DUT opens and closes TCP connections at approximately the same rate.

Measurement TCP Connections Per Second MUST be reported for each test iteration (for each object size).

Test Procedures and Expected Results The test procedure is designed to measure the TCP connections per second rate of the DUT/SUT at the sustaining period of the traffic load profile. The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (Initial connections per second) and meets the test results validation criteria when it was very minimally utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IP types: IPv4 only, IPv6 only, and IPv4 and IPv6 mixed traffic distribution.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial connections per second" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial connections per second" before the sustain phase. The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT continue to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish the target objective ("Target connections per second") defined in . The test equipment MUST follow the traffic load profile definition as described in . During the ramp up and sustain phase of each test iteration, other KPIs such as inspected throughput, concurrent TCP connections, and application transactions per second MUST NOT reach the maximum value the DUT/SUT can support. The test results for specific test iterations MUST NOT be reported as valid results if the above mentioned KPI (especially inspected throughput) reaches the maximum value. (Example: If the test iteration with 64 KByte of HTTP response object size reached the maximum inspected throughput limitation of the DUT/SUT, the test iteration MAY be interrupted and the result for 64 KByte must not be reported.) The test equipment MUST start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective ("Target connections per second") in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable TCP connections per second within the test results validation criteria.

HTTP Throughput

Objective Determine the sustainable inspected throughput of the DUT/SUT for HTTP transactions varying the HTTP response object size.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in Target inspected throughput: Aggregated line rate of the interface(s) used in the DUT/SUT or the value defined based on the requirement for a specific deployment scenario Initial throughput: 10% of "Target inspected throughput" Note: Initial throughput is not a KPI to report. This value is configured on the traffic generator and used to perform Step 1: "Test Initialization and Qualification" described under . Number of HTTP response object requests (transactions) per connection: 10 RECOMMENDED HTTP response object size: 1, 16, 64, 256 KByte, and mixed objects defined in . Mixed Objects

Object size (KByte)	Number of requests/ Weight
0.2	1
6	1
8	1
9	1
10	1
25	1
26	1
35	1
59	1
347	1

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of attempt transactions.
Traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent TCP connections MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. This confirms the DUT opens and closes TCP connections at approximately the same rate.

Measurement Inspected Throughput and HTTP Transactions per Second MUST be reported for each object size.

Test Procedures and Expected Results The test procedure is designed to measure HTTP throughput of the DUT/ SUT. The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (Initial throughput) and meets the test results validation criteria when it was very minimal utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IPv4 and IPv6 traffic distribution and HTTP response object sizes.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial inspected throughput" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial inspected throughput" during the sustain phase. Measure all KPI as defined in . The measured KPIs during the sustain phase MUST meet the test results validation criteria "a" defined in . The test results validation criteria "b" and "c" are OPTIONAL for step 1. If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish the target objective ("Target inspected throughput") defined in . The test equipment MUST start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable inspected throughput within the test results validation criteria and measure the KPI metric Transactions per Second. The final test iteration MUST be performed for the test duration defined in .

HTTP Transaction Latency

Objective Using HTTP traffic, determine the HTTP transaction latency when DUT is running with sustainable HTTP transactions per second supported by the DUT/SUT under different HTTP response object sizes. Test iterations MUST be performed with different HTTP response object sizes in two different scenarios. One with a single transaction and the other with multiple transactions within a single TCP connection. For consistency both the single and multiple transaction test MUST be configured with the same HTTP version Scenario 1: The client MUST negotiate HTTP and close the connection with FIN immediately after the completion of a single transaction (GET and RESPONSE). Scenario 2: The client MUST negotiate HTTP and close the connection FIN immediately after completion of 10 transactions (GET and RESPONSE) within a single TCP connection.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in Target objective for scenario 1: 50% of the connections per second measured in benchmarking test TCP/HTTP Connections Per Second Target objective for scenario 2: 50% of the inspected throughput measured in benchmarking test HTTP Throughput Initial objective for scenario 1: 10% of "Target objective for scenario 1" Initial objective for scenario 2: 10% of "Target objective for scenario 2" Note: The Initial objectives are not a KPI to report. These values are configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under . HTTP transaction per TCP connection: Test scenario 1 with a single transaction and test scenario 2 with 10 transactions. HTTP with GET request requesting a single object. The RECOMMENDED object sizes are 1, 16, and 64 KByte. For each test iteration, the client MUST request a single HTTP response object size.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of attempt transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent TCP connections MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. This confirms the DUT opens and closes TCP connections at approximately the same rate.
After ramp up the DUT MUST achieve the "Target objective" defined in and remain in that state for the entire test duration (sustain phase).

Measurement TTFB (minimum, average, and maximum) and TTLB (minimum, average, and maximum) MUST be reported for each object size.

Test Procedures and Expected Results The test procedure is designed to measure TTFB or TTLB when the DUT/SUT is operating close to 50% of its maximum achievable connections per second or inspected throughput. The test procedure consists of two major steps: Step 1 ensures the DUT/SUT is able to reach the initial performance values and meets the test results validation criteria when it was very minimally utilized. Step 2 measures the latency values within the test results validation criteria. This test procedure MAY be repeated multiple times with different IP types (IPv4 only, IPv6 only, and IPv4 and IPv6 mixed traffic distribution), HTTP response object sizes, and single and multiple transactions per connection scenarios.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial objective" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial objective" before the sustain phase. The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish "Target objective" defined in . The test equipment MUST follow the traffic load profile definition as described in . The test equipment SHOULD start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT MUST reach the desired value of the target objective in the sustain phase. Measure the minimum, average, and maximum values of TTFB and TTLB.

Concurrent TCP/HTTP Connection Capacity

Objective Determine the number of concurrent TCP connections that the DUT/ SUT sustains when using HTTP traffic.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be noted for this benchmarking test:

Client IP address ranges defined in
Server IP address ranges defined in
Traffic distribution ratio between IPv4 and IPv6 defined in
Target concurrent connection: Initial value from product datasheet or the value defined based on the requirement for a specific deployment scenario.
Initial concurrent connection: 10% of "Target concurrent connection" Note: Initial concurrent connection is not a KPI to report. This value is configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under .
Maximum connections per second during ramp up phase: 50% of maximum connections per second measured in benchmarking test TCP/HTTP Connections per second
Ramp up time (in traffic load profile for "Target concurrent connection"): "Target concurrent connection" / "Maximum connections per second during ramp up phase"
Ramp up time (in traffic load profile for "Initial concurrent connection"): "Initial concurrent connection" / "Maximum connections per second during ramp up phase"

The client MUST negotiate HTTP and each client MAY open multiple concurrent TCP connections per server endpoint IP. Each client sends 10 GET requests requesting 1 KByte HTTP response object in the same TCP connection (10 transactions/TCP connection) and the delay (think time) between each transaction MUST be X seconds. X = ("Ramp up time" + "steady state time") /10 The established connections MUST remain open until the ramp down phase of the test. During the ramp down phase, all connections MUST be successfully closed with FIN.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of total attempted transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).

Measurement Average Concurrent TCP Connections MUST be reported for this benchmarking test.

Test Procedures and Expected Results The test procedure is designed to measure the concurrent TCP connection capacity of the DUT/SUT at the sustaining period of the traffic load profile. The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (Initial concurrent connection) and meets the test results validation criteria when it was very minimally utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IPv4 and IPv6 traffic distributions.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure test equipment to establish "Initial concurrent TCP connections" defined in . Except ramp up time, the traffic load profile MUST be defined as described in . During the sustain phase, the DUT/SUT MUST reach the "Initial concurrent TCP connections". The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish the target objective ("Target concurrent TCP connections"). The test equipment MUST follow the traffic load profile definition (except ramp up time) as described in . During the ramp up and sustain phase, the other KPIs such as inspected throughput, TCP connections per second, and application transactions per second MUST NOT reach the maximum value the DUT/SUT can support. The test equipment MUST start to measure and record KPIs defined in . Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable concurrent TCP connections capacity within the test results validation criteria.

TCP/QUIC Connections per Second with HTTPS Traffic

Objective Using HTTPS traffic, determine the sustainable TLS session establishment rate supported by the DUT/SUT under different throughput load conditions. Test iterations MUST include common cipher suites and key strengths as well as forward looking stronger keys. Specific test iterations MUST include ciphers and keys defined in . For each cipher suite and key strengths, test iterations MUST use a single HTTPS response object size defined in to measure connections per second performance under a variety of DUT/SUT security inspection load conditions.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in Target connections per second: Initial value from product datasheet or the value defined based on the requirement for a specific deployment scenario. Initial connections per second: 10% of "Target connections per second" (Note: Initial connections per second is not a KPI to report. This value is configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under .) RECOMMENDED ciphers and keys defined in The client MUST negotiate HTTPS and close the connection without error immediately after the completion of one transaction. In each test iteration, the client MUST send a GET request requesting a fixed HTTPS response object size. The RECOMMENDED object sizes are 1, 2, 4, 16, and 64 KByte.

Test Results Validation Criteria The following criteria are the test results validation criteria. The test results validation criteria MUST be monitored during the whole test duration.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of attempt transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections. If HTTP/3 is used, the number of terminated QUIC connections due to unexpected errors MUST be less than 0.001% (1 out of 100,000 connections) of total initiated QUIC connections.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent TCP connections generation rate MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. If HTTP/3 is used, the concurrent QUIC connections generation rate MUST be constant during steady state and any deviation of concurrent QUIC connections MUST be less than 10%. This confirms the DUT opens and closes connections at approximately the same rate.

Measurement If HTTP 1.1 or HTTP/2 is used, TCP connections per second MUST be reported for each test iteration (for each object size). If HTTP/3 is used, QUIC connections per second MUST be measured and reported for each test iteration (for each object size). The KPI metric TLS Handshake Rate can be measured in the test using 1 KByte object size.

Test Procedures and Expected Results The test procedure is designed to measure the TCP or QUIC connections per second rate of the DUT/SUT at the sustaining period of the traffic load profile. The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (Initial connections per second) and meets the test results validation criteria when it was very minimally utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IPv4 and IPv6 traffic distributions.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial connections per second" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial connections per second" before the sustain phase. The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish "Target connections per second" as defined in . The test equipment MUST follow the traffic load profile definition as described in . During the ramp up and sustain phase, other KPIs such as inspected throughput, concurrent TCP/QUIC connections, and application transactions per second MUST NOT reach the maximum value the DUT/SUT can support. The test results for the specific test iteration MUST NOT be reported as valid results, if the above mentioned KPI (especially inspected throughput) reaches the maximum value. (Example: If the test iteration with 64 KByte of HTTPS response object size reached the maximum inspected throughput limitation of the DUT, the test iteration MAY be interrupted, and the result for 64 KByte should not be reported). The test equipment MUST start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective ("Target connections per second") in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable connections per second within the test results validation criteria.

HTTPS Throughput

Objective Determine the sustainable inspected throughput of the DUT/SUT for HTTPS transactions varying the HTTPS response object size. Test iterations MUST include common cipher suites and key strengths as well as forward looking stronger keys. Specific test iterations MUST include the ciphers and keys defined in .

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in Target inspected throughput: Aggregated line rate of the interface(s) used in the DUT/SUT or the value defined based on the requirement for a specific deployment scenario. Initial throughput: 10% of "Target inspected throughput" Note: Initial throughput is not a KPI to report. This value is configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under . Number of HTTPS response object requests (transactions) per connection: 10 RECOMMENDED ciphers and keys defined in RECOMMENDED HTTPS response object size: 1, 16, 64, 256 KByte, and mixed objects defined in under .

Number of failed Application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of attempt transactions.
Traffic MUST be generated at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent generated TCP connections MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. If HTTP/3 is used, the concurrent generated QUIC connections MUST be constant during steady state and any deviation of concurrent QUIC connections MUST be less than 10%. This confirms the DUT opens and closes connections at approximately the same rate.

Measurement Inspected Throughput and HTTPS Transactions per Second MUST be reported for each object size.

Test Procedures and Expected Results The test procedure consists of three major steps: Step 1 ensures the DUT/SUT is able to reach the performance value (Initial throughput) and meets the test results validation criteria when it was very minimally utilized. Step 2 determines whether the DUT/SUT is able to reach the target performance value within the test results validation criteria. Step 3 determines the maximum achievable performance value within the test results validation criteria. This test procedure MAY be repeated multiple times with different IPv4 and IPv6 traffic distribution and HTTPS response object sizes.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial throughput" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial throughput" during the sustain phase. Measure all KPI as defined in . The measured KPIs during the sustain phase MUST meet the test results validation criteria "a" defined in . The test results validation criteria "b", and "c" are OPTIONAL for step 1. If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 3: Test Iteration Determine the achievable average inspected throughput within the test results validation criteria. The final test iteration MUST be performed for the test duration defined in .

HTTPS Transaction Latency

Objective Using HTTPS traffic, determine the HTTPS transaction latency when DUT/SUT is running with sustainable HTTPS transactions per second supported by the DUT/SUT under different HTTPS response object sizes. Scenario 1: The client MUST negotiate HTTPS and close the connection immediately after the completion of a single transaction (GET and RESPONSE). Scenario 2: The client MUST negotiate HTTPS and close the connection immediately after completion of 10 transactions (GET and RESPONSE) within a single TCP or QUIC connection.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test: Client IP address ranges defined in Server IP address ranges defined in Traffic distribution ratio between IPv4 and IPv6 defined in RECOMMENDED cipher suites and key sizes defined in Target objective for scenario 1: 50% of the connections per second measured in benchmarking test TCP/QUIC Connections per Second with HTTPS Traffic Target objective for scenario 2: 50% of the inspected throughput measured in benchmarking test HTTPS Throughput Initial objective for scenario 1: 10% of "Target objective for scenario 1" Initial objective for scenario 2: 10% of "Target objective for scenario 2" Note: The Initial objectives are not a KPI to report. These values are configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under . HTTPS transaction per TCP or QUIC connection: Test scenario 1 with a single transaction and scenario 2 with 10 transactions HTTPS with GET request requesting a single object. The RECOMMENDED object sizes are 1, 16, and 64 KByte. For each test iteration, the client MUST request a single HTTPS response object size.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of attempt transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections. If HTTP/3 is used, the number of terminated QUIC connections due to unexpected errors MUST be less than 0.001% (1 out of 100,000 connections) of total initiated QUIC connections.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
Concurrent TCP or QUIC connections MUST be constant during steady state and any deviation of concurrent TCP connections MUST be less than 10%. If HTTP/3 is used, the concurrent generated QUIC connections MUST be constant during steady state and any deviation of concurrent QUIC connections MUST be less than 10%. This confirms the DUT opens and closes connections at approximately the same rate.
After ramp up the DUT/SUT MUST achieve the "Target objective" defined in the parameter and remain in that state for the entire test duration (sustain phase).

Measurement TTFB (minimum, average, and maximum) and TTLB (minimum, average, and maximum) MUST be reported for each object size.

Test Procedures and Expected Results The test procedure is designed to measure TTFB or TTLB when the DUT/SUT is operating close to 50% of its maximum achievable connections per second or inspected throughput. The test procedure consists of two major steps: Step 1 ensures the DUT/SUT is able to reach the initial performance values and meets the test results validation criteria when it was very minimally utilized. Step 2 measures the latency values within the test results validation criteria. This test procedure MAY be repeated multiple times with different IP types (IPv4 only, IPv6 only, and IPv4 and IPv6 mixed traffic distribution), HTTPS response object sizes, and single, and multiple transactions per connection scenarios.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure the traffic load profile of the test equipment to establish "Initial objective" as defined in . The traffic load profile MUST be defined as described in . The DUT/SUT MUST reach the "Initial objective" before the sustain phase. The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish the "Target objective" defined in . The test equipment MUST follow the traffic load profile definition as described in . The test equipment MUST start to measure and record all specified KPIs. Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT MUST reach the desired value of the target objective in the sustain phase. Measure the minimum, average, and maximum values of TTFB and TTLB.

Concurrent TCP/QUIC Connection Capacity with HTTPS Traffic

Objective Determine the number of concurrent TCP/QUIC connections the DUT/SUT sustains when using HTTPS traffic.

Test Setup Testbed setup MUST be configured as defined in . Any specific testbed configuration changes (number of interfaces and interface type, etc.) MUST be documented.

Test Parameters In this section, benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT parameters MUST conform to the requirements defined in . Any configuration changes for this specific benchmarking test MUST be documented.

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The following parameters MUST be documented for this benchmarking test:

Client IP address ranges defined in
Server IP address ranges defined in
Traffic distribution ratio between IPv4 and IPv6 defined in
RECOMMENDED cipher suites and key sizes defined in
Target concurrent connections: Initial value from product datasheet or the value defined based on the requirement for a specific deployment scenario.
Initial concurrent connections: 10% of "Target concurrent connections" Note: Initial concurrent connection is not a KPI to report. This value is configured on the traffic generator and used to perform Step1: "Test Initialization and Qualification" described under .
Connections per second during ramp up phase: 50% of maximum connections per second measured in benchmarking test TCP/QUIC Connections per second with HTTPS Traffic
Ramp up time (in traffic load profile for "Target concurrent connections"): "Target concurrent connections" / "Maximum connections per second during ramp up phase"
Ramp up time (in traffic load profile for "Initial concurrent connections"): "Initial concurrent connections" / "Maximum connections per second during ramp up phase"

The client MUST perform HTTPS transactions with persistence and each client can open multiple concurrent connections per server endpoint IP. Each client sends 10 GET requests requesting 1 KByte HTTPS response objects in the same TCP/QUIC connections (10 transactions/connection) and the delay (think time) between each transaction MUST be X seconds. X = ("Ramp up time" + "steady state time") /10 The established connections MUST remain open until the ramp down phase of the test. During the ramp down phase, all connections MUST be successfully closed with FIN.

Number of failed application transactions (receiving any HTTP response code other than 200 OK) MUST be less than 0.001% (1 out of 100,000 transactions) of total attempted transactions.
Number of terminated TCP connections due to unexpected TCP RST sent by DUT/SUT MUST be less than 0.001% (1 out of 100,000 connections) of total initiated TCP connections. If HTTP/3 is used, the number of terminated QUIC connections due to unexpected errors MUST be less than 0.001% (1 out of 100,000 connections) of total initiated QUIC connections
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).

Measurement Average Concurrent TCP or QUIC Connections MUST be reported for this benchmarking test.

Step 1: Test Initialization and Qualification Verify the link status of all connected physical interfaces. All interfaces are expected to be in "UP" status. Configure test equipment to establish "Initial concurrent TCP connections" defined in . Except ramp up time, the traffic load profile MUST be defined as described in . During the sustain phase, the DUT/SUT MUST reach the "Initial concurrent connections". The measured KPIs during the sustain phase MUST meet the test results validation criteria "a", and "b" defined in . If the KPI metrics do not meet the test results validation criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: Test Run with Target Objective Configure test equipment to establish the target objective ("Target concurrent connections"). The test equipment MUST follow the traffic load profile definition (except ramp up time) as described in . During the ramp up and sustain phase, the other KPIs such as inspected throughput, TCP or QUIC connections per second, and application transactions per second MUST NOT reach the maximum value that the DUT/SUT can support. The test equipment MUST start to measure and record KPIs defined in . Continue the test until all traffic profile phases are completed. Within the test results validation criteria, the DUT/SUT is expected to reach the desired value of the target objective in the sustain phase. Follow step 3, if the measured value does not meet the target value or does not fulfill the test results validation criteria.

Step 3: Test Iteration Determine the achievable concurrent TCP/QUIC connections within the test results validation criteria.

IANA Considerations This document makes no specific request of IANA. The IANA has assigned IPv4 and IPv6 address blocks in that have been registered for special purposes. The IPv6 address block 2001:2::/48 has been allocated for the purpose of IPv6 Benchmarking and the IPv4 address block 198.18.0.0/15 has been allocated for the purpose of IPv4 Benchmarking . This assignment was made to minimize the chance of conflict in case a testing device were to be accidentally connected to the part of the Internet.

Security Considerations The primary goal of this document is to provide benchmarking terminology and methodology for next-generation network security devices for use in a laboratory isolated test environment. However, readers should be aware that there is some overlap between performance and security issues. Specifically, the optimal configuration for network security device performance may not be the most secure, and vice-versa. Testing security platforms with working exploits and malware carries risks. Ensure proper access controls are implemented to prevent unintended exposure to vulnerable networks or systems. The cipher suites recommended in this document are for test purposes only. The cipher suite recommendation for a real deployment is outside the scope of this document. Security assessment of an NGFW/NGIPS product could also include an analysis whether any type of uncommon traffic characteristics would have a significant impact on performance. Such performance impacts would allow an attacker to use such specifically crafted traffic as a DoS attack to reduce the remaining performance available to other traffic through the NGFW/NGIPS. Such uncommon traffic characteristics might include for example IP fragmented traffic, specific type of application traffic, or uncommonly high HTTP transaction rate traffic.

Contributors The following individuals contributed significantly to the creation of this document: Alex Samonte, Amritam Putatunda, Aria Eslambolchizadeh, Chao Guo, Chris Brown, Cory Ford, David DeSanto, Jurrie Van Den Breekel, Michelle Rhines, Mike Jack, Ryan Liles, Samaresh Nair, Stephen Goudreault, Tim Carlin, and Tim Otto.

Acknowledgements The authors wish to acknowledge the members of NetSecOPEN for their participation in the creation of this document. Additionally, the following members need to be acknowledged: Anand Vijayan, Chris Marshall, Jay Lindenauer, Michael Shannon, Mike Deichman, Ryan Riese, and Toulnay Orkun.

References Normative References Key words for use in RFCs to Indicate Requirement Levels In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements. Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings. Informative References Benchmarking Methodology for Firewall Performance This document discusses and defines a number of tests that may be used to describe the performance characteristics of firewalls. In addition to defining the tests, this document also describes specific formats for reporting the results of the tests. This document is a product of the Benchmarking Methodology Working Group (BMWG) of the Internet Engineering Task Force (IETF). This memo provides information for the Internet community. Applicability Statement for RFC 2544: Use on Production Networks Considered Harmful The Benchmarking Methodology Working Group (BMWG) has been developing key performance metrics and laboratory test methods since 1990, and continues this work at present. The methods described in RFC 2544 are intended to generate traffic that overloads network device resources in order to assess their capacity. Overload of shared resources would likely be harmful to user traffic performance on a production network, and there are further negative consequences identified with production application of the methods. This memo clarifies the scope of RFC 2544 and other IETF BMWG benchmarking work for isolated test environments only, and it encourages new standards activity for measurement methods applicable outside that scope. This document is not an Internet Standards Track specification; it is published for informational purposes. Benchmarking Terminology for Firewall Performance This document defines terms used in measuring the performance of firewalls. It extends the terminology already used for benchmarking routers and switches with definitions specific to firewalls. [STANDARDS-TRACK] Benchmarking Methodology for Network Interconnect Devices This document is a republication of RFC 1944 correcting the values for the IP addresses which were assigned to be used as the default addresses for networking test equipment. This memo provides information for the Internet community. IPv6 Benchmarking Methodology for Network Interconnect Devices The benchmarking methodologies defined in RFC 2544 are IP version independent. However, RFC 2544 does not address some of the specificities of IPv6. This document provides additional benchmarking guidelines, which in conjunction with RFC 2544, lead to a more complete and realistic evaluation of the IPv6 performance of network interconnect devices. IPv6 transition mechanisms are outside the scope of this document. This memo provides information for the Internet community. Special-Purpose IP Address Registries This memo reiterates the assignment of an IPv4 address block (192.0.0.0/24) to IANA. It also instructs IANA to restructure its IPv4 and IPv6 Special-Purpose Address Registries. Upon restructuring, the aforementioned registries will record all special-purpose address blocks, maintaining a common set of information regarding each address block. Internet Protocol, Version 6 (IPv6) Specification This document specifies version 6 of the Internet Protocol (IPv6). It obsoletes RFC 2460. The Transport Layer Security (TLS) Protocol Version 1.3 This document specifies version 1.3 of the Transport Layer Security (TLS) protocol. TLS allows client/server applications to communicate over the Internet in a way that is designed to prevent eavesdropping, tampering, and message forgery. This document updates RFCs 5705 and 6066, and obsoletes RFCs 5077, 5246, and 6961. This document also specifies new requirements for TLS 1.2 implementations. QUIC: A UDP-Based Multiplexed and Secure Transport This document defines the core of the QUIC transport protocol. QUIC provides applications with flow-controlled streams for structured communication, low-latency connection establishment, and network path migration. QUIC includes security measures that ensure confidentiality, integrity, and availability in a range of deployment circumstances. Accompanying documents describe the integration of TLS for key negotiation, loss detection, and an exemplary congestion control algorithm. Using TLS to Secure QUIC This document describes how Transport Layer Security (TLS) is used to secure QUIC. QUIC Loss Detection and Congestion Control This document describes loss detection and congestion control mechanisms for QUIC. HTTP/2 This specification describes an optimized expression of the semantics of the Hypertext Transfer Protocol (HTTP), referred to as HTTP version 2 (HTTP/2). HTTP/2 enables a more efficient use of network resources and a reduced latency by introducing field compression and allowing multiple concurrent exchanges on the same connection. This document obsoletes RFCs 7540 and 8740. HTTP/3 The QUIC transport protocol has several features that are desirable in a transport for HTTP, such as stream multiplexing, per-stream flow control, and low-latency connection establishment. This document describes a mapping of HTTP semantics over QUIC. This document also identifies HTTP/2 features that are subsumed by QUIC and describes how HTTP/2 extensions can be ported to HTTP/3. Transmission Control Protocol (TCP) This document specifies the Transmission Control Protocol (TCP). TCP is an important transport-layer protocol in the Internet protocol stack, and it has continuously evolved over decades of use and growth of the Internet. Over this time, a number of changes have been made to TCP as it was specified in RFC 793, though these have only been documented in a piecemeal fashion. This document collects and brings those changes together with the protocol specification from RFC 793. This document obsoletes RFC 793, as well as RFCs 879, 2873, 6093, 6429, 6528, and 6691 that updated parts of RFC 793. It updates RFCs 1011 and 1122, and it should be considered as a replacement for the portions of those documents dealing with TCP requirements. It also updates RFC 5961 by adding a small clarification in reset handling while in the SYN-RECEIVED state. The TCP header control bits from RFC 793 have also been updated based on RFC 3168. QPACK: Field Compression for HTTP/3 This specification defines QPACK: a compression format for efficiently representing HTTP fields that is to be used in HTTP/3. This is a variation of HPACK compression that seeks to reduce head-of-line blocking. <author/> <date/> </front> </reference> <reference anchor="fastly" target="https://www.fastly.com/blog/measuring-quic-vs-tcp-computational-efficiency"> <front> <title>Can QUIC match TCP's computational efficiency? An in depth overview of HTTP/2

Test Methodology - Security Effectiveness Evaluation

Test Objective This test methodology verifies the DUT/SUT is able to detect, prevent, and report the vulnerabilities. In this test, background test traffic will be generated to utilize the DUT/SUT. In parallel, a number of malicious traffic will be sent to the DUT/SUT as encrypted and as well as clear text payload formats using a traffic generator. defines the selection of the malicious traffic from the Common Vulnerabilities and Exposures (CVE) list for testing. The following KPIs are measured in this test:

Number of blocked CVEs
Number of bypassed (nonblocked) CVEs
Background traffic performance (verify if the background traffic is impacted while sending CVE toward DUT/SUT)
Accuracy of DUT/SUT statistics in terms of vulnerabilities reporting

Testbed Setup The same testbed MUST be used for security effectiveness tests and as well as for benchmarking test cases defined in .

Test Parameters In this section, the benchmarking test specific parameters are defined.

DUT/SUT Configuration Parameters DUT/SUT configuration parameters MUST conform to the requirements defined in . The same DUT configuration MUST be used for the security effectiveness test and as well as for benchmarking test cases defined in . The DUT/SUT MUST be configured in inline mode and all detected attack traffic MUST be dropped and the session MUST be reset

Test Equipment Configuration Parameters Test equipment configuration parameters MUST conform to the requirements defined in . The same client and server IP ranges MUST be configured as used in the benchmarking test cases. In addition, the following parameters MUST be documented for this benchmarking test:

Background Traffic: 45% of maximum HTTP throughput and 45% of Maximum HTTPS throughput supported by the DUT/SUT (measured with object size 64 KByte in the benchmarking tests "HTTP(S) Throughput" defined in and ).
RECOMMENDED CVE traffic transmission Rate: 10 CVEs per second
It is RECOMMENDED to generate each CVE multiple times (sequentially) at 10 CVEs per second
Ciphers and keys for the encrypted CVE traffic MUST use the same cipher configured for HTTPS traffic related benchmarking tests ( - )

Test Results Validation Criteria The following criteria are the test results validation criteria. The test results validation criteria MUST be monitored during the whole test duration.

Number of failed application transactions in the background traffic MUST be less than 0.01% of attempted transactions.
Number of terminated TCP or QUIC connections of the background traffic (due to unexpected errors) MUST be less than 0.01% of total initiated TCP connections in the background traffic.
During the sustain phase, traffic MUST be forwarded at a constant rate (considered as a constant rate if any deviation of traffic forwarding rate is less than 5%).
False positive MUST NOT occur in the background traffic.

Measurement Following KPI metrics MUST be reported for this test scenario: Mandatory KPIs:

Blocked CVEs: It MUST be represented in the following ways:
- Number of blocked CVEs out of total CVEs
- Percentage of blocked CVEs
Unblocked CVEs: It MUST be represented in the following ways:
- Number of unblocked CVEs out of total CVEs
- Percentage of unblocked CVEs
Background traffic behavior: It MUST be represented in one of the followings ways:
- No impact: Considered as "no impact'" if any deviation of traffic forwarding rate is less than or equal to 5 % (constant rate)
- Minor impact: Considered as "minor impact" if any deviation of traffic forwarding rate is greater than 5% and less than or equal to10% (i.e. small spikes)
- Heavily impacted: Considered as "Heavily impacted" if any deviation of traffic forwarding rate is greater than 10% (i.e. large spikes) or reduced the background HTTP(S) throughput greater than 10%
DUT/SUT reporting accuracy: DUT/SUT MUST report all detected vulnerabilities.

Optional KPIs:

List of unblocked CVEs

Test Procedures and Expected Results The test procedure is designed to measure the security effectiveness of the DUT/SUT at the sustaining period of the traffic load profile. The test procedure consists of two major steps. This test procedure MAY be repeated multiple times with different IPv4 and IPv6 traffic distributions.

Step 1: Background Traffic Generate background traffic at the transmission rate defined in . The DUT/SUT MUST reach the target objective (HTTP(S) throughput) in sustain phase. The measured KPIs during the sustain phase MUST meet all the test results validation criteria defined in . If the KPI metrics do not meet the acceptance criteria, the test procedure MUST NOT be continued to "Step 2".

Step 2: CVE Emulation While generating background traffic (in sustain phase), send the CVE traffic as defined in the parameter section. The test equipment MUST start to measure and record all specified KPIs. Continue the test until all CVEs are sent. The measured KPIs MUST meet all the test results validation criteria defined in . In addition, the DUT/SUT should either report the detected vulnerabilities in the log correctly or if, for example, a different naming convention is used, there MUST be reference material available that will allow for verification that the correct vulnerability was detected. This reference material MUST be cited in the report.

DUT/SUT Classification This document aims to classify the DUT/SUT into four different categories based on its maximum supported firewall throughput performance number defined in the vendor datasheet. This classification MAY help users to determine specific configuration scales (e.g., number of ACL entries), traffic profiles, and attack traffic profiles, scaling those proportionally to DUT/SUT sizing category. The four different categories are Extra Small (XS), Small (S), Medium (M), and Large (L). The RECOMMENDED throughput values for the following categories are: Extra Small (XS) - Supported throughput less than or equal to1Gbit/s Small (S) - Supported throughput greater than 1Gbit/s and less than or equal to 5Gbit/s Medium (M) - Supported throughput greater than 5Gbit/s and less than or equal to 10Gbit/s Large (L) - Supported throughput greater than 10Gbit/s