Arista Networks introduced advanced capabilities to maximize AI cluster performance and efficiency.
Cluster Load Balancing (CLB) in Arista EOS® maximizes AI workload performance with consistent, low-latency network flows, while Arista CloudVision® Universal Network Observability™ (CV UNO™) now offers AI job-centric observability for enhanced troubleshooting and rapid issue inference ensuring job completion reliability at scale.
The Arista EOS Smart AI Suite is designed for AI-grade robustness and protection and empowers AI clusters with an innovation called Cluster Load Balancing — a new Ethernet-based AI load balancing solution based on RDMA queue pairs that enables high bandwidth utilization between spines and leaves. AI clusters usually have low quantities of large bandwidth flows. Basic load balancing methods are often inefficient for AI workloads, resulting in uneven traffic distribution and increased tail latency. CLB addresses this by using RDMA-aware flow placement, to ensure uniform high performance for all flows while keeping tail latency low. CLB takes a global approach, optimizing traffic flow in both directions, leaf-to-spine and spine-to-leaf, ensuring balanced utilization and consistent low latency.
CV UNO, the AI-driven 3600 Network Observability platform powered by Arista AVA™, delivers seamless, end-to-end AI job visibility by unifying network, system, and AI job data within the Arista Network Data Lake (NetDL™). EOS NetDL Streamer, a real-time telemetry framework that continuously streams granular network data from Arista switches into NetDL. Unlike traditional SNMP polling, which relies on periodic queries and can miss critical updates, the EOS NetDL Streamer provides low-latency, high-frequency, event-driven insights into network performance, key to supercharging large-scale AI training and inferencing infrastructure. Designed for AI accelerator clusters, it accelerates impact analysis, pinpoints issues with precision, and enables rapid resolution—ensuring job completion times are minimized.
Some of the key benefits include:
- AI Job Monitoring – Unlocks a comprehensive view of AI job health metrics, including job completion times, congestion indicators (ECN-marked packets, PFC pause frames, packet drops), and buffer/link utilization for real-time insights.
- Deep-Dive Analytics – Uncovers critical job-specific insights by analyzing network devices, server NICs (e.g., PFC out-of-sync events, RDMA errors, PCIe fatal errors), and associated flows — pinpointing performance bottlenecks with precision.
- Flow Visualization – Harnesses the power of CV topology mapping to gain real-time, intuitive visibility into AI job flows at microsecond granularity — accelerating issue inference and resolution.
- Proactive Resolution – Detects anomalies early and correlates network and compute performance within NetDL — ensuring uninterrupted, high-efficiency AI workload execution.
Arista’s Etherlink™ AI Platforms deliver ultra-high-performance, standards-based Ethernet systems for next-gen AI networks. Offering 800G/400G fixed, modular, and distributed platforms that are forward-compatible with Ultra Ethernet Consortium (UEC), Etherlink scales from small AI clusters to massive deployments with 100,000+ accelerators. Arista features the AI Analyzer, powered by Arista AVA, which delivers high-resolution traffic data at 100-microsecond intervals, enabling precise performance optimization and troubleshooting. This allows network administrators to optimize performance, quickly troubleshoot issues, and make informed decisions for AI-driven networks. Arista AVA also powers a remote EOS AI Agent, that streams telemetry from SuperNICs or servers to NetDL, ensuring seamless network monitoring, debugging, and QoS consistency across the entire stack.
CLB is available today on 7260X3, 7280R3, 7500R3 and 7800R3 platforms. Support on 7060X6 and 7060X5 platforms scheduled for Q2 2025. Support for 7800R4 scheduled for 2H 2025.
CV UNO is available today. The observability enhancements for AI are in active customer trials, with general availability scheduled for Q2 2025.
The Latest
A recent Rocket Software and Foundry study found that just 28% of organizations fully leverage their mainframe data, a concerning statistic given its critical role in powering AI models, predictive analytics, and informed decision-making ...
What kind of ROI is your organization seeing on its technology investments? If your answer is "it's complicated," you're not alone. According to a recent study conducted by Apptio ... there is a disconnect between enterprise technology spending and organizations' ability to measure the results ...
In today’s data and AI driven world, enterprises across industries are utilizing AI to invent new business models, reimagine business and achieve efficiency in operations. However, enterprises may face challenges like flawed or biased AI decisions, sensitive data breaches and rising regulatory risks ...
In MEAN TIME TO INSIGHT Episode 12, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses purchasing new network observability solutions....
There's an image problem with mobile app security. While it's critical for highly regulated industries like financial services, it is often overlooked in others. This usually comes down to development priorities, which typically fall into three categories: user experience, app performance, and app security. When dealing with finite resources such as time, shifting priorities, and team skill sets, engineering teams often have to prioritize one over the others. Usually, security is the odd man out ...
IT outages, caused by poor-quality software updates, are no longer rare incidents but rather frequent occurrences, directly impacting over half of US consumers. According to the 2024 Software Failure Sentiment Report from Harness, many now equate these failures to critical public health crises ...
In just a few months, Google will again head to Washington DC and meet with the government for a two-week remedy trial to cement the fate of what happens to Chrome and its search business in the face of ongoing antitrust court case(s). Or, Google may proactively decide to make changes, putting the power in its hands to outline a suitable remedy. Regardless of the outcome, one thing is sure: there will be far more implications for AI than just a shift in Google's Search business ...
In today's fast-paced digital world, Application Performance Monitoring (APM) is crucial for maintaining the health of an organization's digital ecosystem. However, the complexities of modern IT environments, including distributed architectures, hybrid clouds, and dynamic workloads, present significant challenges ... This blog explores the challenges of implementing application performance monitoring (APM) and offers strategies for overcoming them ...
Service disruptions remain a critical concern for IT and business executives, with 88% of respondents saying they believe another major incident will occur in the next 12 months, according to a study from PagerDuty ...
IT infrastructure (on-premises, cloud, or hybrid) is becoming larger and more complex. IT management tools need data to drive better decision making and more process automation to complement manual intervention by IT staff. That is why smart organizations invest in the systems and strategies needed to make their IT infrastructure more resilient in the event of disruption, and why many are turning to application performance monitoring (APM) in conjunction with high availability (HA) clusters ...