Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
MLPerf™ Inference v2.1 with NVIDIA GPU-Based Benchmarks on Dell PowerEdge Servers
Introduction What is new in MLPerf Inference v2.1? Growth of MLPerf Inference as a benchmark Dell results Notable results Highlighting performance scores System performance details PowerEdge server, GPU, and Software specifications Performance improvements Conclusion and Key Takeaways
Summary
Acknowledgment References

Summary

Home > AI Solutions > Artificial Intelligence > White Papers > MLPerf™ Inference v2.1 with NVIDIA GPU-Based Benchmarks on Dell PowerEdge Servers > Summary

Summary

  • Key takeaways include:

    • Dell PowerEdge R750xa, XE8545, and XR12 servers with NVIDIA accelerators deliver outstanding performance for different tasks such as image classification, object detection, natural language processing, recommender systems, image segmentation, and speech recognition.
    • Combining Dell servers with NVIDIA accelerators is an excellent choice to obtain a high-performance inference system. The PowerEdge XE8545 server with the NVIDIA A100 SXM 80 GB card yielded the highest performance for different classes of workloads. Other PCIE-based servers, such as the PowerEdge R750xa server with NVIDIA A100 and the PowerEdge XR12 server with NVIDIA A2 accelerators are also an excellent choice for deep learning inferencing.
    • Our submissions include performance and power results. PowerEdge XR12 and XE8545 servers deliver outstanding performance/watt and are aptly suited for inference workloads.
    • The Power Edge XR12 server delivers high energy-efficient performance for different suites such as the data center and edge.
    • Modern inference deployment can come with multiple competing objectives; datapoints, software from benchmarking tools such as MLPerf can serve as a good starting point to aid customers use cases around those challenges.
    • Customers can also refer to previous rounds of MLPerf inference results such as v0.7, v1.0, v1.1 and v2.0 to see an evolution in performance offered by our submissions, which include variety of NVIDIA accelerators and Dell PowerEdge Servers.