Right now, we’re exploring how Ethernet stacks up towards InfiniBand in AI/ML environments, specializing in how Cisco Silicon One™ manages community congestion and enhances efficiency for AI/ML workloads. This publish emphasizes the significance of benchmarking and KPI metrics in evaluating community options, showcasing the Cisco Zeus Cluster outfitted with 128 NVIDIA® H100 GPUs and cutting-edge congestion administration applied sciences like dynamic load balancing and packet spray.

Networking requirements to satisfy the wants of AI/ML workloads

AI/ML coaching workloads generate repetitive micro-congestion, stressing community buffers considerably. The east-to-west GPU-to-GPU visitors throughout mannequin coaching calls for a low-latency, lossless community cloth. InfiniBand has been a dominant expertise within the high-performance computing (HPC) setting and currently within the AI/ML setting.

Ethernet is a mature different, with superior options that may handle the rigorous calls for of the AI/ML coaching workloads and Cisco Silicon One can successfully execute load balancing and handle congestion. We got down to benchmark and evaluate Cisco Silicon One versus NVIDIA Spectrum-X™ and InfiniBand.

Analysis of community cloth options for AI/ML

Community visitors patterns differ based mostly on mannequin measurement, structure, and parallelization strategies utilized in accelerated coaching. To guage AI/ML community cloth options, we recognized related benchmarks and key efficiency indicator (KPI) metrics for each AI/ML workload and infrastructure groups, as a result of they view efficiency by totally different lenses.

We established complete assessments to measure efficiency and generate metrics particular to AI/ML workload and infrastructure groups. For these assessments, we used the Zeus Cluster, that includes devoted backend and storage with a typical 3-stage leaf-spine Clos cloth community, constructed with Cisco Silicon One–based mostly platforms and 128 NVIDIA H100 GPUs. (See Determine 1.)

We developed benchmarking suites utilizing open-source and industry-standard instruments contributed by NVIDIA and others. Our benchmarking suites included the next (see additionally Desk 1):

• Distant Direct Reminiscence Entry (RDMA) benchmarks—constructed utilizing IBPerf utilities—to guage community efficiency throughout congestion created by incast
• NVIDIA Collective Communication Library (NCCL) benchmarks, which consider software throughput throughout coaching and inference communication section amongst GPUs
• MLCommons MLPerf set of benchmarks, which evaluates essentially the most understood metrics, job completion time (JCT) and tokens per second by the workload groups

Legend:

JCT = Job Completion Time

Bus BW = Bus bandwidth

ECN/PFC = Express Congestion Notification and Precedence Circulate Management

NCCL benchmarking towards congestion avoidance options

Congestion builds up throughout the again propagation stage of the coaching course of, the place a gradient sync is required amongst all of the GPUs taking part in coaching. Because the mannequin measurement will increase, so does the gradient measurement and the variety of GPUs. This creates large micro-congestion within the community cloth. Determine 2 reveals outcomes of the JCT and visitors distribution benchmarking. Word how Cisco Silicon One helps a set of superior options for congestion avoidance, akin to dynamic load balancing (DLB) and packet spray strategies, and Knowledge Middle Quantized Congestion Notification (DCQCN) for congestion administration.

Determine 2 illustrates how the NCCL benchmarks stack up towards totally different congestion avoidance options. We examined the commonest collectives with a number of totally different message sizes to spotlight these metrics. The outcomes present that JCT improves with DLB and packet spray for All-to-All, which causes essentially the most congestion because of the nature of communication. Though JCT is essentially the most understood metric from an software’s perspective, JCT doesn’t present how successfully the community is utilized—one thing the infrastructure crew must know. This information might assist them to:

• Enhance the community utilization to get higher JCT
• Know what number of workloads can share the community cloth with out adversely impacting JCT
• Plan for capability as use circumstances improve

To gauge community cloth utilization, we calculated Jain’s Equity Index, the place LinkTxᵢ is the quantity of transmitted visitors on cloth hyperlink:

The index worth ranges from 0.0 to 1.0, with increased values being higher. A worth of 1.0 represents the proper distribution. The Site visitors Distribution on Cloth Hyperlinks chart in Determine 2 reveals how DLB and packet spray algorithms create a near-perfect Jain’s Equity Index, so visitors distribution throughout the community cloth is nearly good. ECMP makes use of static hashing, and relying on stream entropy, it might probably result in visitors polarization, inflicting micro-congestion and negatively affecting JCT.

Silicon One versus NVIDIA Spectrum-X and InfiniBand

The NCCL Benchmark – Aggressive Evaluation (Determine 3) reveals how Cisco Silicon One performs towards NVIDIA Spectrum-X and InfiniBand applied sciences. The info for NVIDIA was taken from the SemiAnalysis publication. Word that Cisco doesn’t know the way these assessments have been carried out, however we do know that the cluster measurement and GPU to community cloth connectivity is much like the Cisco Zeus Cluster.

Bus Bandwidth (Bus BW) benchmarks the efficiency of collective communication by measuring the velocity of operations involving a number of GPUs. Every collective has a particular mathematical equation reported throughout benchmarking. Determine 3 reveals that Cisco Silicon One – All Cut back performs comparably to NVIDIA Spectrum-X and InfiniBand throughout numerous message sizes.

Community cloth efficiency evaluation

The IBPerf Benchmark compares RDMA efficiency towards ECMP, DLB, and packet spray, that are essential for assessing community cloth efficiency. Incast eventualities, the place a number of GPUs ship knowledge to at least one GPU, usually trigger congestion. We simulated these circumstances utilizing IBPerf instruments.

Determine 4 reveals how Aggregated Session Throughput and JCT reply to totally different congestion avoidance algorithms: ECMP, DLB, and packet spray. DLB and packet spray attain Hyperlink Bandwidth, enhancing JCT. It additionally illustrates how DCQCN handles micro-congestions, with PFC and ECN ratios enhancing with DLB and considerably dropping with packet spray. Though JCT improves barely from DLB to packet spray, the ECN ratio drops dramatically on account of packet spray’s splendid visitors distribution.

Coaching and inference benchmark

The MLPerf Benchmark – Coaching and Inference, revealed by the MLCommons group, goals to allow truthful comparability of AI/ML programs and options.

We centered on AI/ML knowledge heart options by executing coaching and inference benchmarks. To attain optimum outcomes, we extensively tuned throughout compute, storage, and networking elements utilizing congestion administration options of Cisco Silicon One. Determine 5 reveals comparable efficiency throughout numerous platform distributors. Cisco Silicon One with Ethernet performs like different vendor options for Ethernet.

Conclusion

Our deep dive into Ethernet and InfiniBand inside AI/ML environments highlights the outstanding prowess of Cisco Silicon One in tackling congestion and boosting efficiency. These modern developments showcase the unwavering dedication of Cisco to offer strong, high-performance networking options that meet the rigorous calls for of at present’s AI/ML functions.

Many because of Vijay Tapaskar, Will Eatherton, and Kevin Wollenweber for his or her assist on this benchmarking course of.

Discover safe AI infrastructure

Uncover safe, scalable, and high-performance AI infrastructure it’s essential to develop, deploy, and handle AI workloads securely if you select Cisco Safe AI Manufacturing unit with NVIDIA.

Share: