]> Huawei Technologies

Riesstrasse 25 Munich 80992 Germany zhe.lou@huawei.com

Huawei Technologies France S.A.S.U.

18, Quai du Point du Jour Boulogne-Billancourt 92100 France luigi.iannone@huawei.com

Huawei Technologies

Beiqing Road, Haidian District Beijing 100095 China zhouyujing3@huawei.com

Huawei Technologies

Huawei base in Bantian, Longgang District Shenzhen China zhangcuimin@huawei.com

SFC Internet-Draft This memo introduces "Signaling In-Network Computing operations" (SINC), a mechanism to enable in-packet operation signaling for in-network computing for specific scenarios like NetReduce, NetDistributedLock, NetSequencer, etc. In particular, this solution allows to flexibly communicate computation parameters to be used in conjunction with the packets' payload, to signal to in-network SINC-enabled devices the computing operations to be performed.

Introduction According to the original design, the Internet performs just "store and forward" of packets, and leaves more complex operations at the end-points. However, new emerging applications could benefit from in-network packet processing to improve the overall system efficiency (, ). The formation of the COIN Research Group in IRTF encourages people to explore this emerging technology and its impact on the Internet architecture. The "Use Cases for In-Network Computing" draft introduces some use cases to demonstrate how real applications can benefit from COIN and show essential requirements demanded by COIN applications. Recent research has shown that network devices undertake some computing tasks can greatly improve the network and application performance in some scenarios like aggregating path-computing , key-value(K-V) cache , and strong consistency . Their implementations are mainly based on programmable network devices, by using P4 or other languages. In the context of such heterogeneity of scenarios, it is desirable to have a generic and flexible protocol to explicitly signal the computing operation to be performed by network devices, which is applicable to many use cases, enabling easier deployment of these research results. This document specifies a signaling architecture for in-network computing operation. The computing functions are hosted on network devices, which can be perceived as network SINC service instances. It focuses on the design of the data plane, while the control plane will be depicted in a separate draft. Service Function Chaining (SFC) is used as a running example on how to tunnel the SINC header to the in-network device and implement the desired in-network computation. Nevertheless, the mechanism can be adapted to other transport protocols, like Remote Direct Memory Access (RDMA) , but such adaptation is out of the scope of this document.

Requirements Language The key words "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 and when, and only when, they appear in all capitals, as shown here.

Terminology This document uses the terms as defined in , and . This document assume that the reader is familiar with the Service Function Chaining architecture.

SINC Relevant Use Cases Hereafter a few relevant use cases are described, namely NetReduce, NetDistributedLock, and Net Sequencer, in order to help understanding of the requirements for a general framework. Such a framework, should be generic enough to accommodate a large variety of use cases, besides the ones described in this document.

NetReduce Over the last decade, the rapid development of the Deep Neural Networks (DNN) has greatly improved the performance of many Artificial Intelligence (AI) applications like computer vision and natural language processing. However, DNN training is a computation intensive and time consuming task, which has been increased exponentially (computation time gets doubled every 3.4 months ) in the past 10 years. Scale-up techniques concentrating on the computing capability of a single device cannot meet the expectation. Distributed DNN training approaches with synchronous data parallelism like Parameter Server and All-Reduce are commonly employed in practice, which on the other hand, become increasingly a network-bound workload since communication becomes a bottleneck at scale (,). Comparing with the host oriented solutions, in-network aggregation approaches like SwithML and SHARP could potentially reduce nearly half of the bandwidth needed for data aggregation by offloading gradients aggregation from the host to the network switch. The SwitchML solution uses UDP for network transport. The system solely relies on application layer logic to trigger retransmission for packet loss, which leads to extra latency and reduces the training performance. The SHARP solution on the contrary, uses Remote Direct Memory Access (RDMA) to provide reliable transmission . As the Infini-Band (IB) technology requires specific hardware support, this solution is not very cost-effective. NetReduce doesn't depend on dedicated hardware and provides a general in-network aggregation solution that is suitable for Ethernet networks.

NetDistributedLock In the majority of distributed system, the lock primitive is a widely used concurrency control mechanism. For large distributed systems, there is commonly a dedicated lock manager that nodes compete to gain read and/or write permissions of a resource. The lock manager is often abstracted as Compare And Swap (CAS) or Fetch Add (FA) operations. The lock manager is typically running on a server, causing a limitation on the performance by the speed of disk I/O transaction. When the load increases, for instance in the case of database transactions processed on a single node, the lock manager becomes a major performance bottleneck, consuming nearly 75% of transaction time . The multi-node distributed lock processing superimposes the communication latency between nodes, which makes the performance even worse. Therefore offloading the lock manager function from the server to the network switch might be a much better choice, as the switch is capable of managing lock function efficiently. Meanwhile it releases the server for other computation tasks. The test results in NetLock show that the lock manager running on a switch is able to answer 100 million requests per second, nearly 10 times more than what a lock server can do.

NetSequencer Transaction managers are centralized solutions to the consistency issue for distributed transactions, such as GTM in Postgre-XL (, ). However, as a centralized module, transaction managers have became a bottleneck in large scale high-performance distributed systems.
The work introduces a server based networked sequencer, which is a kind of task manager assigning monotonically increasing sequence number for transactions. In , the authors shows that the maximum throughput is 122 Million requests per second (Mrps), at the cost of an increased average latency. This bounded throughput will impact the scalability of distributed systems. Meanwhile, the authors also test the bottlenecks for varies optimization methods, including CPU, DMA bandwidth and PCIe RTT, which is introduced by the CPU centric architecture. For a programmable switch, a sequencer is a rather simple operation to implement, while the pipeline architecture can avoid bottlenecks. It is worth trying to implement a switch based sequencer, which set the performance goals as hundreds of Mrps and latency in the order of microseconds.

Simple Generic Operations The COIN use case draft illustrates some general requirements for scenarios like in-network control and distributed AI, where the aforementioned use cases belong to. One of the requirements defined in is that any in-network computing system must provide means to specify the constraints for placing execution logic in certain logical execution points (and their associated physical locations). In case of NetReduce, NetDistributedLock and NetSequencer, data aggregation, lock management and sequence number generation functions can be offloaded respectively onto the network switch. We can see that those functions are based on some "simple" and "generic" operators, as shown in . Programmable switches are capable of performing those basic operations by executing one or more operators, without impacting the forwarding performance (, ). Example of in-network operators.

Use Case	Operation	Description
NetReduce	Sum value (SUM)	The network device sums the collected parameters together and outputs the resulting value.
NetLock	Compare And Swap or Fetch-and-Add (CAS or FA)	By comparing the request value with the status of its own lock, the network device sends out whether the host has the acquired the lock. Through the CAS and FA, host can implement shared and exclusive locks.
NetSequencer	Fetch-and-Add (FA)	The network device offers a counter service and provides a monotonically increasing sequence number for the host.

SINC Overview This section describes the various elements and functional modules in the SINC system and explains how they work together. The SINC computing protocol and extensions are designed for limited domains such as the data center network instead of across the Internet. The requirements and semantics are specifically limited, as defined in the previous sections. The main deployment model is to place SINC-capable switches/routers, aiming to take over part of the data computing operations during the data transmission. For instance, in the case of NetLock, Top-of-Rack switches can be equipped with SINC capabilities to manage I/O locks. In the case of NetReduce, SINC-capable switches can be deployed in a centric point where all data has to pass through, to achieve on-path aggregation/reduction. shows the architecture of a SINC network. In the computing service chain, a host sends out packets containing data operations to be executed in the network. The data operation description should be carried in the packet itself by using the SINC header. Once the packet is in the SINC domain, it includes a SINC header, so that SINC-enabled switches and router have access to such header and can perform the desired operation directly on the in-network device. Note that hosts can also be SINC enabled, in that case the proxies are not necessary.

SINC Architecture. |->|Service|->|-->| Egress | | Proxy | | +-------+ | | Proxy | +---------+ +-------------+ +---------+ ]]>

SINC Header The SINC header, has a fixed length of 16 octets and it is appended right after the Service Path Header, carries the data operation information, used for on-path in-switch SFs.

SINC Context Header.

• Reserved: Flags field reserved for future use. MUST be set to zero on transmission and ignored on reception.
• Loopback flag (L): Zero (0) indicates that the packet should be sent to the destination after the data operation. One (1) indicates that the packet should be sent back to the source node after the data operation.
• Group ID: The group ID identifies different groups. Each group is associated with one task.
• Number of Data Sources: Total number of data source nodes that are part of the group.
• Data Source ID: Unique identifier of the data source node of the packet.
• Sequence Number (SeqNum): The SeqNum is used to identify different requests within one group.
• Data Operation: The operation to be executed by the SF (see ).
• Data Offset: The in-packet offset from the SINC context header to the data required by the operation.

SFC for Signal In-Network Computing As previously stated, Service Function Chaining (SFC) is used as a running example on how to tunnel the SINC header to the in-network device and implement the desired in-network computation. shows the architecture of a SFC-based SINC network. In the computing service chain, a host sends out packets containing data operations to be executed in the network. The data operation description should be carried in the packet itself by using a SINC-specific NSH encapsulation. Once the SINC packet is in the SFC domain, the Service Function Forwarder (SFF) is responsible for forwarding packets to one or more connected service functions according to information carried in the SFC encapsulation. The Service Function (SF) is responsible for implementing data operations.

SINC for SFC Architecture. | SFF |-->| | SFF | |-->| SFF |-->| SFC Proxy | +-----------+ +-----+ | +-----+ | +-----+ +-----------+ | | | | +-----+ | | | SF | | | +-----+ | +-----------+ ]]>

SFC Elements As shown in , the SFC proxy, SFF, and SINC switch/router containing SFF and SF, are used. The SFC proxy is required to support SFC-unaware hosts to encapsulate the packets with correct NSH header and SINC context header, and to forward the packets to a correct SFF. The SFF forwards packets based on the Service Path Header (SPH), as specified in . The SFC-unaware hosts can only add the SINC information in the payload after the transport layer encapsulation. The SFC proxy needs to associate packets to a group and, hence, to a specific operation to be done in-network. For TCP and UDP packets, the five-tuple is sufficient for flow identification. For RoCEv2 packets, the destination port number is set to 4791 for the indication of the InfiniBand Base Transport Header (IB BTH), which cannot be used for flow identification. Therefore, a combination of source IP address, destination IP address, and Destination Queue Pair number should be used to for flow identification. For packets from the SFC-unaware hosts that requires SINC operation, the ingress SFC proxy will copy the SINC information to a SINC context header and set the Data Offset value accordingly ((see )). Based on the Group ID, the SPI is matched and the NSH based header is built. With a SFC encapsulation, the SINC packet will be forwarded to SFF. The egress SFC proxy removes the NSH header, including the SINC context header, before forwarding the packets to destination. With the standardized context header, the SFs can be decoupled from transport layer encapsulation. The SFs perform the data operation as defined in the headers, update the original payload with the results, and forward the packets to the next hop.

SINC NSH encapsulation This section defines the SINC header fields as part of the NSH encapsulation for SFC .

NSH Base Header The SINC NSH header is basically another type of NSH MD header. SINC NSH encapsulation uses the NSH Meta Data (MD) fixed-length context headers to carry the data operation information. Please refer to the NSH for a detailed SFC basic header description. This draft suggest the base header specifies MD type = 0x4, to allow a fixed length context header immediately following the service path header.

NSH Base Header, where "MD Type is set to 0x4.

NSH Service Path Header Following the NSH basic header there is the Service Path Header, show in , as defined in .

NSH Service Path Header.

Complete SINC NSH Header By stacking the previously shown headers, the complete SINC NSH header, meaning the NSH base header, NSH Service Path Header, and the SINC Header, all together are shown in .

SINC NSH Header.

SFC-based SINC Workflow This section describes the SINC system workflow, focusing on elements and key information changes through the workflow. Since SINC's use-cases will use a programmable switch to host the SF, it is assumed that both SFF and SF are colocated on the same switch, as shown in .

An Example of SINC system. | SFF |-->| | SFF | |-->| SFF |-->| SFC Proxy | +-----------+ +-----+ | +-----+ | +-----+ +-----------+ | | | | +-----+ | | | SF | | | +-----+ | +-----------+ ]]>

For the sake of clarity, a simple example with one sender (Host A) and one Receiver (Host B) is provided. Packet processing goes through the following steps:

1. Host A transmit the packet containing data that can be processed by the SF on the switch.
2. The SFC Proxy performs encapsulates the original packet with the NSH header containing the SINC Header. Based on the information obtained from control plane, the SFC proxy builds a SINC context header pre-pended to the original packet. The SFC proxy encapsulates the packet as the transport protocol indicated by the SFC.
3. SFF forwards the SINC packets to the specified SF. As shown in , when the packet reaches the SINC switch, the packet reaches the egress point of the tunnel and the header is removed. The SFF looks up the SPI table and SI table and forwards the packet to the SF.
4. SF performs the Computing Operation according to the content of the SINC header. The SF verifies the Group ID and Data Source ID in the SINC context header, then preforms the required computing according to the Data Operation field. When the computing is done, the payload is replaced with the result. The packet is re-encapsulated with the NSH SINC header. The SI is reduce by 1 while other fields are untouched. Then, the packet is forwarded to the SFC Egress.
5. Packets are forwarded to Host B, its the final destination. When the packet reaches the SFC Egress, it looks up the SPI table and SI table and realizes it is the egress. It removes the NSH encapsulation and forwards the inner packet to the final destination.

SINC Control Plane SINC networks need to deploy and control the whole life-cycle of the task. It should be able to manage the full life-cycle from the initialization to the end of the computing task and give support to the computing tasks. The detailed design of the control plane will be discussed in a separate document.

Security Considerations In-network computing exposes computing data to network devices, which inevitably raises security and privacy considerations. The security problems faced by in-network computing include, but are not limited to:

• Trustworthiness of participating devices
• Data hijacking and tampering
• Private data exposure

This documents assume that the deployment is done in a trusted environment. For example, in a data center network or a private network. A fine security analysis will be provided in future revisions of this memo.

IANA Considerations This document defines a new NSH fixed length context header. As such, IANA is requested to add the entry depicted in , to the "NSH MD Types" sub registry of the "Network Service Header Parameter" registry. [Note to RFC Editor: If IANA assign a different value the authors will update the document accordingly] NSH MD type allocation for SINC NSH Context Header.

MD Type	Description	Reference
0x4	NSA SINC MD Header	[This Document]

Acknowledgements Dirk Trossen's feedback was of great help in improving this document.

References Normative References This document describes an architecture for the specification, creation, and ongoing maintenance of Service Function Chains (SFCs) in a network. It includes architectural concepts, principles, and components used in the construction of composite services through deployment of SFCs, with a focus on those to be standardized in the IETF. This document does not propose solutions, protocols, or extensions to existing protocols. 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. 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. This document provides an overview of the issues associated with the deployment of service functions (such as firewalls, load balancers, etc.) in large-scale environments. The term "service function chaining" is used to describe the definition and instantiation of an ordered list of instances of such service functions, and the subsequent "steering" of traffic flows through those service functions. The set of enabled service function chains reflects operator service offerings and is designed in conjunction with application delivery and service and network policy. This document also identifies several key areas that the Service Function Chaining (SFC) working group will investigate to guide its architectural and protocol work and associated documents. This document describes a Network Service Header (NSH) imposed on packets or frames to realize Service Function Paths (SFPs). The NSH also provides a mechanism for metadata exchange along the instantiated service paths. The NSH is the Service Function Chaining (SFC) encapsulation required to support the SFC architecture (defined in RFC 7665). Informative References n.d. n.d. n.d. School of Computer Science, China University of Geosciences,Wuhan,China New Jersey Institute of Technology (NJIT),USA Concordia University,Montreal,Canada Intel,USA University of Bologna,Italy RWTH Aachen University RWTH Aachen University Huawei Technologies Duesseldorf GmbH Concordia University InterDigital Communications, LLC University College London University College London Computing in the Network (COIN) comes with the prospect of deploying processing functionality on networking devices, such as switches and network interface cards. While such functionality can be beneficial in several contexts, it has to be carefully placed into the context of the general Internet communication. This document discusses some use cases to demonstrate how real applications can benefit from COIN and to showcase essential requirements that have to be fulfilled by COIN applications.

Computing Capability Operation abstraction Computing tasks and application are becoming increasingly complex. The complexities are caused by model extension. If some computing tasks are directly offloaded on network devices, the universality of devices will be reduced. Complex models can be disassembled into basic calculation operation, such as addition, subtraction, Max, etc. Therefore, a more appropriate offloading method is to disassemble complex tasks into basic computing operations. The DOIN Network needs to provide a set of general computing abilities abstraction framework. The application, management and computing network nodes can negotiate and calculate resources according to the abstract computing abilities. For each calculation operation, such as addition, subtraction and maximization, the corresponding settings should be found in the abstract scheme and the abstraction should be realized. The abstraction of computing abilities represents that network nodes should give the same output with the same input and operation. The example of DOIN Operation

OpName	Operation Explanation
Max	Maximum value of several parameters
MIN	Minimum value
SUM	Sum value
PROD	Product value
LAND	Logical and
BAND	Bit-wise and
LOR	Logical or
BOR	Bit-wise or
LXOR	Logical xor
BXOR	Bit-wise xor
WRITE	Write value accord to key
READ	Read value accord to key
DELETE	Delete value accord to key
CAS	Compare and swap. compare the value of the key and old value. If not same, swap old value to key value. Return old key value.
CAADD	Compare and add. compare the value of the key and expected value. If same, add add-value to key value. Return old key value.
CASUB	Compare and subtract. compare the value of the key and expected value. If same, sub sub-value to key value. Return old key value.
FA	Fetch and add. Fetch value according key. Add add-value to key value. Return old key-value.
FASUB	Fetch and subtract.Fetch value according key. Subtract sub-value to key value. Return old key value.
FAOR	Fetch and OR. Fetch value according key. Key value get logical or operation with parameter. Return old key value.
FAADD	Fetch and ADD. Fetch value according key. Key value get logical add operation with parameter. Return old key value.
FANAND	Fetch and NAND. Fetch value according key. Key value get logical NAND operation with parameter. Return old key value.
FAXOR	Fetch and XOR. Fetch value according key. Key value get logical XOR operation with parameter. Return old key value.

Defining an appropriate abstract model of computing capability is helpful for interoperability between computing devices. They are also a necessary condition for the application and practice of In-Network computing technology. Most of the existing papers are based on a single computing task, and corresponding private protocols are proposed. The lack of unified protocols makes the equipment complex and unstable. It also makes the research task of In-Network computing impossible to disassemble. For example, scholars who study hardware prefer to focus on optimizing the processing efficiency of a single operator in the device, but they are not good at the message protocol with the design operator. The computing capability abstraction model should support a variety of operators, including the possibility of operator extension.