HPE sets new AI inference world records: Continued excellence in performance

[ComputeExperts]

These benchmark achievements highlight the versatility and robustness of HPE's AI solutions, and underscore our focus on delivering best-in-class security, performance, and efficiency across a broad spectrum of AI applications.

HPE is thrilled to announce that we have achieved numerous world records with another round of benchmark testing using the latest MLPerf Inference: Datacenter v5.0 benchmark suite. This announcement from the MLCommons Association solidifies our position as a leader in AI innovation.

We recently introduced the newest generation of HPE ProLiant Compute servers, bringing a new level of security, performance, and efficiency to IT operations. We are excited to share that two of these servers, the HPE ProLiant Compute DL380a Gen12 and the HPE ProLiant Compute DL384 Gen12, played a pivotal role in our latest round of benchmark testing.

HPE ProLiant Compute DL380a Gen12 server

In this testing, we achieved top results in over 23 different scenarios, a significant increase from results previously accomplished. This remarkable improvement underscores the exceptional capabilities of our AI-optimized systems and our commitment to continuous innovation. A key factor, of course, behind these impressive results was the utilization of the latest NVIDIA accelerator technology.

Benchmark #1s increase by two: leading across multiple tests

In October 2024, HPE achieved top ranking in a diverse array of AI inference scenarios, including the renowned benchmarks for SDXL, DLRM-v2-99, Retinanet, and Llama2-70B. Additionally – with the latest test results – we can proudly report that HPE also secured the top position in two more benchmark tests, as follows:

• Text generation Mixture of Experts (MoE): Mixtral-8x7B. A versatile MoE model designed to generate text for question answering, math, and code generation tasks better than some large language models.
• Text summarization Large Language Model: GPT-J 99. A large-scale language model known for its ability to generate coherent and contextually relevant text, making it ideal for applications such as text generation and summarization.

These achievements highlight the versatility and robustness of our AI solutions, and underscore our focus on delivering best-in-class security, performance, and efficiency across a broad spectrum of AI applications.

HPE ProLiant Compute DL384 Gen12 serverThe future of AI is promising

HPE continues to innovate and lead the charge in AI as represented by our latest benchmark achievements. We are continuously exploring new technologies, partnerships, and innovations to ensure our customers remain at the forefront of AI advancements, as demonstrated by the exceptional capabilities of the HPE ProLiant Compute DL384 Gen12 and the HPE ProLiant DL380a Gen12 servers, part of NVIDIA AI Computing by HPE. This collaboration between HPE and NVIDIA was created to deliver co-developed enterprise AI solutions and joint go-to-market integrations that assist businesses in streamlining the development and deployment of AI applications – from pilot to production.

For more detailed information about our test results, explore the MLCommons results interactive table.

If you want to harness these powerful AI servers for your enterprise, please visit our HPE ProLiant Compute DL384 Gen12 and HPE ProLiant Compute DL380a Gen12 web pages.

Meet HPE blogger Sonja Hickey.

Sonja Hickey is currently a product marketing manager at Hewlett Packard Enterprise (HPE), focusing on HPE ProLiant servers. Sonja has extensive marketing and product management experience having worked in the IT industry since 1997 with enterprise IT companies that include Dell, Sun Microsystems and Zebra Technologies.

In 2011, Sonja co-authored a book, IT Operations Management, which explores best practices associated with IT infrastructure management, especially related to cloud and virtualized environments.  Sonja’s education includes an MBA from the University of Chicago’s Booth School, as well as a MS and BS in Engineering from the University of Illinois in Champaign-Urbana.

Compute Experts
Hewlett Packard Enterprise

twitter.com/hpe_compute
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/servers