Dirk Kutscher

Personal web page

Towards a Unified Transport Protocol for In-Network Computing in Support of RPC-based Applications

without comments

The emerging term In-Network Computin (INC) [inc] in particular refers applying on-path programmable networking devices (e.g., switches and routers between clients and servers) as an accelerator or function offloader to boost throughput, reduce server load, or improve latency, typically in a well-controlled data center network environment.

Some INC implementations evolved from programmable data plane systems and align with the trend of network programmability at large. In recent year, it has been shown to support many promising applications (e.g., caching, aggregation, and agreement). For example, in distributed machine learning (DML), training nodes produce data (gradients) that needs to be aggregated or reduced -- and the result could be distributed to one or multiple consumers. As another example, the NetClone system [netclone] uses in-network forwarder to replicate RPC invocation messages and to perform more informed forwarding based on observed latencies for accelerating RPC communication.

While it is possible to achieve this kind of operation purely with end-to-end communication between worker nodes, performance can be dramatically improved by offloading both the operation processing and the data dissemination to nodes in the network. These in-network processors are often conceived as semi-transparent performance enhancing on-path elements, i.e., they are not the actual endpoints in transport protocol sessions and would intercept packets with application data and potentially generate new data that they would have to transmit.

In our Internet Draft draft-song-inc-transport-protocol-req-01.txt, we are discussing this problem and are formulating some requirements for the design of future transport protocols in this space.


Written by dkutscher

January 25th, 2024 at 7:02 am