Optimal GPU and Network Coordinated Performance

Arista Networks (NYSE: ANET) today announced a technology demonstration of AI Data Centers in order to align compute and network domains as a single managed AI entity, in collaboration with NVIDIA. In order to build optimal generative AI networks with lower job completion times, customers can configure, manage, and monitor AI clusters uniformly across key building blocks including networks, NICs, and servers. This demonstrates the first step in achieving a multi-vendor, interoperable ecosystem that enables control and coordination between AI networking and AI compute infrastructure.

Need for Uniform Controls

As the size of AI clusters and large language models (LLMs) grows, the complexity and sheer volume of disparate parts of the puzzle grow apace. GPUs, NICs, switches, optics, and cables must all work together to form a holistic network. Customers need uniform controls between their AI servers hosting NICs and GPUs, and the AI network switches at different tiers. All these elements are reliant upon each other for proper AI job completion but operate independently. This could lead to misconfiguration or misalignment between aspects of the overall ecosystem, such as between NICs and the switch network, which can dramatically impact job completion time, since network issues can be very difficult to diagnose. Large AI clusters also require coordinated congestion management to avoid packet drops and under-utilization of GPUs, as well as coordinated management and monitoring to optimize compute and network resources in tandem.

Introducing the Arista AI Agent

At the heart of this solution is an Arista EOS-based agent enabling the network and the host to communicate with each other and coordinate configurations to optimize AI clusters. Using a remote AI agent, EOS running on Arista switches can be extended to directly-attached NICs and servers to allow a single point of control and visibility across an AI Data Center as a holistic solution. This remote AI agent, hosted directly on an NVIDIA BlueField-3 SuperNIC or running on the server and collecting telemetry from the SuperNIC, allows EOS, on the network switch, to configure, monitor, and debug network problems on the server, ensuring end-to-end network configuration and QoS consistency. AI clusters can now be managed and optimized as a single homogenous solution.

“Arista aims to improve efficiency of communication between the discovered network and GPU topology to improve job completion times through coordinated orchestration, configuration, validation, and monitoring of NVIDIA accelerated compute, NVIDIA SuperNICs, and Arista network infrastructure,” said John McCool, Chief Platform Officer for Arista Networks.

End-to-End AI Communication and Optimization

This new technology demonstration highlights how an Arista EOS-based remote AI agent allows the combined, interdependent AI cluster to be managed as a single solution. EOS running in the network can now be extended to servers or SuperNICs via remote AI agents to enable instantaneous tracking and reporting of performance degradation or failures between hosts and networks, so they can be rapidly isolated and the impact minimized. Since EOS-based network switches are constantly aware of accurate network topology, extending EOS down to SuperNICs and servers with the remote AI agent further enables coordinated optimization of end-to-end QoS between all elements in the AI Data Center to reduce job completion time.

“Best-of-breed Arista networking platforms with NVIDIA’s compute platforms and SuperNICs enable coordinated AI Data Centers. The new ability to extend Arista’s EOS operating system with remote AI agents on hosts promises to solve a critical customer problem of AI clusters at scale, by delivering a single point of control and visibility to manage AI availability and performance as a holistic solution,” said Zeus Kerravala, Principal Analyst at ZK Research.

Arista will demonstrate the AI agent technology at the Arista IPO 10th anniversary celebration in NYSE on June 5th, with customer trials expected in 2H 2024.

If you are a member of the analyst or financial community and are interested in attending the NYSE event, please register here.

To read more about the new AI Centers check out CEO and Chairperson Jayshree Ullal’s blog here.

About Arista

Arista Networks is an industry leader in data-driven, client-to-cloud networking for large data center/AI, campus, and routing environments. Its award-winning platforms deliver availability, agility, automation, analytics, and security through an advanced network operating stack. For more information, visit www.arista.com.

ARISTA and EOS are among the registered and unregistered trademarks of Arista Networks, Inc. in jurisdictions worldwide. Other company names or product names may be trademarks of their respective owners. Additional information and resources can be found at www.arista.com. This press release contains forward-looking statements including, but not limited to, statements regarding the performance and capabilities of Arista’s products and services. All statements other than statements of historical fact are statements that could be deemed forward-looking statements. Forward-looking statements are subject to risks and uncertainties that could cause actual performance or results to differ materially from those expressed in the forward-looking statements, including rapid technological and market change, customer requirements and industry standards, as well as other risks stated in our filings with the SEC available on Arista's website at www.arista.com and the SEC's website at www.sec.gov. Arista disclaims any obligation to publicly update or revise any forward-looking statement to reflect events that occur or circumstances that exist after the date on which they were made.

View source version on businesswire.com: https://www.businesswire.com/news/home/20240529365397/en/