This March, at NVIDIA GTC, HPE and NVIDIA unveiled a number of new enterprise AI solutions from NVIDIA AI Computing by HPE. NVIDIA AI Computing by HPE accelerates time to value with a comprehensive portfolio of co-developed AI solutions to streamline the development and deployment of AI models. Enterprises of all sizes can harness a complete AI portfolio to either train, tune or inference their models at the edge, in a dedicated AI data center or colocation facility.
Let's take a closer look at what was announced for HPE ProLiant Compute:
- Support for NVIDIA HGX B300 NVL16 with HPE ProLiant Compute XD series
- HPE ProLiant Compute DL380a Gen12 with NVIDIA RTX™ PRO 6000 Blackwell Server Edition
- HPE ProLiant Compute DL384b Gen12 with NVIDIA GB200 NVL4
Next-level security with HPE ProLiant Compute Gen12 Servers
HPE Integrated Lights-Out (iLO) safeguards every phase of the server lifecycle with industry-leading HPE iLO Silicon Root of Trust. The latest HPE ProLiant Compute Gen12 portfolio sets a new standard for enterprise security with built-in safeguards at every layer – from the chip to the cloud. HPE iLO 7 introduces an enhanced and dedicated security processor called secure enclave that establishes an unbreakable chain of trust to protect against firmware attacks and creates full line-of-sight from the factory and throughout HPE’s trusted supply chain. HPE iLO 7 delivers the first server with post-quantum cryptography and meets the requirements for FIPS 140-3 Level 3 certification[i], a high-level cryptographic security standard.
Support for NVIDIA HGX B300 NVL16 with HPE ProLiant Compute XD series
HPE ProLiant Compute XD servers will support NVIDIA HGX B300 NVL16 which includes 16 NVIDIA Blackwell Ultra GPUs for customers looking to train and fine-tune large AI models.
For those seeking ultra-scalable acceleration, the new HPE ProLiant Compute DL380a Gen12 is the answer. Highlights include support for the new RTX PRO 6000 Blackwell GPU, expanded capacity for ten double-wide or sixteen single-wide GPUs along with direct liquid cooling option for enhanced performance and efficiency. The HPE ProLiant Compute DL380a Gen12 is a purpose-built 4U rack server purpose-built for AI fine tuning and inferencing. The DL380a Gen12 also features PCIe Gen5 and DDR5 memory, providing up to 4TB of RAM and NVMe & EDSFF storage options.
RTX PRO 6000 Blackwell is the ultimate data center GPU for AI and visual computing, delivering breakthrough acceleration for the most demanding enterprise workloads, from agentic and physical AI to scientific computing, graphics, and video applications. Optimized for workloads requiring the compute density and scale that deploying in the data center offers, it features a passively cooled thermal design and 96 GB of ultra-fast GDDR7 memory.
With fifth-generation Tensor Cores for AI-enhanced computing, RTX PRO 6000 Blackwell enables the creation of complex AI agents that can perceive, reason, and act in dynamic environments. By leveraging the power of AI and deep learning, RTX PRO 6000 Blackwell accelerates agentic AI workloads, allowing researchers and developers to build and deploy autonomous systems that can learn, adapt, and interact with their surroundings in real time. With fifth-gen Tensor Cores and support for the second-generation Transformer Engine, RTX PRO 6000 Blackwell delivers up to 5X higher performance than the previous-generation NVIDIA L40S for agentic and generative AI applications.
Offering immense compute power, memory, and throughput for critical AI applications across industries such as healthcare, manufacturing, geoscience, retail, media, and live broadcast. The DL380a Gen12 with RTX PRO 6000 Blackwell enhances performance, scalability, and reliability, driving innovation and pushing boundaries.
HPE ProLiant Compute DL380a Gen12 is available today and the NVIDIA RTX PRO 6000 Blackwell Server Edition PCIe-based GPUs is expected to be available later this year. Learn more.
The HPE ProLiant Compute DL384b Gen12 with the NVIDIA GB200 Grace Blackwell NVL4 Superchip, is engineered to deliver next-level performance for mixed, memory-intensive, and AI workloads. This new rack optimized server is equipped with the NVIDIA GB200 NVL4, which provides fast, unified, and coherent memory, making it ideal for AI fine-tuning and inference tasks using Retrieval Augmented Generation (RAG).
Available in air-cooled or with direct liquid cooling (DLC) it will be the first HPE ProLiant server with the NVIDIA Grace CPU to offer DLC for superior cooling efficiency to improve performance and deliver cost savings by reducing the need for energy-intensive air conditioning
The NVIDIA GB200 NVL4 features 768 GB of HBM3E memory across its four NVIDIA B200 GPUs, providing a combined memory bandwidth of 32 TB/s, and 960 GB of LPDDR5X memory for the NVIDIA Grace CPUs. It utilizes fifth generation NVIDIA NVLink(™) chip-to-chip interconnect to enable high-speed communication between the CPUs and GPUs up to 1.8TB/s.
The HPE ProLiant DL384b Gen12 with its advanced AI capabilities, enables enterprises to deploy large language models for AI fine-tuning and inference, significantly enhancing computing capabilities and accelerating the shift to AI applications.
HPE ProLiant Compute DL384b Gen12 with NVIDIA GB200 NVL4 is expected to be generally available in the calendar Q4 2025. The HPE ProLiant Compute DL384 Gen12 with NVIDIA GH200 NVL2 is available now.
Learn more.
New enhancements for HPE Cloud AI
HPE Private Cloud AI, a first-of-its-kind solution provides the deepest integration to date of NVIDIA AI computing, networking, and software with HPE’s AI storage, compute and the HPE GreenLake cloud. The turnkey offer enables enterprises of every size to gain a fast, efficient, and flexible path for developing and deploying generative, agentic, and physical AI applications.
It is complemented by NVIDIA Spectrum-X Ethernet networking and NVIDIA AI Enterprise software to support accelerated computing use cases such as agentic and generative AI.
NVIDIA AI Enterprise is the cloud-native software platform for the development and deployment of production-grade, end-to-end generative AI pipelines. With NVIDIA AI Enterprise Software, enterprise organizations can build data flywheels for the next era of agentic AI and optimize model performance with enterprise-grade security, support, and stability, ensuring a smooth transition from prototype to production.
NVIDIA AI Computing by HPE continues to deliver the latest AI compute servers to support the full gamut of model development and deployment from training, tuning and inferencing with advanced security with HPE Integrated Lights Out (iLO) and liquid cooling expertise.
To learn more about how HPE ProLiant Compute latest AI servers can accelerate your AI initiatives, please visit HPE ProLiant Compute.
Meet HPE Blogger John Carter.
John received his MBA from the University of Iowa, Henry B. Tippie College of Business, and his BBA from the University of Iowa. He currently resides in Seattle, WA. He is married and has two dogs, a German Shepherd and a Bernese Mountain dog.
Compute Experts
Hewlett Packard Enterprise
twitter.com/hpe_compute
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/servers
[i] As of January 27th, 2025.
ComputeExperts
Our team of Hewlett Packard Enterprise server experts helps you to dive deep into relevant infrastructure topics.
