Hewlett Packard Enterprise (HPE) updated its edge-focused ProLiant platform with a focus on supporting the growing surge in edge-based AI inferencing use cases.

The biggest update is HPE’s new ProLiant Compute EL2000 chassis that is designed to operate in resource-constrained environments. It’s based on the Intel Xeon 6 processor platform for deployment in edge locations and can support up to two EL220 Gen12 servers or a single EL240 Gen12 server, with that latter option also available with Nvidia’s RTX Pro 4500 or 6000 Blackwell Server Edition graphic processing units (GPUs).

These platforms are set for availability later this year.

John Carter, VP for mainstream compute at HPE, noted in a press briefing that HPE has not yet seen telecom operators adopt the Nvidia option.

“They’re starting to explore it,” Carter said. “I don’t think we’ve seen a huge uptick is this part of the kind of workload. Now, other parts of their environment, absolutely, but not yet. They’re looking at ways that they could accelerate, whether that’s within Nvidia or another type of accelerator from another vendor, but I think that’s heavily still in that exploration phase.”

That “exploration phase” includes Comcast’s recent admission that it was working to deploy Nvidia GPUs at its regional facilities to support latency and speed sensitive edge services through what it calls an “AI grid.” That work included a use case where it deployed a small business concierge agent using Personal AI’s small language model (SLM) and a memory platform deployed on HPE’s ProLiant servers.

Updated servers and a ‘ruggedized’ kit

HPE also updated its ProLiant DL145 Gen11 server, which will now include the upcoming AMD EPYC 8005-series (Sorano) processors. That server is a two-rack unit (2U) system supporting up to 84 cores and will have a version validated to use the Nvidia RTX Pro 4500 Blackwell Server Edition GPU for edge AI inferencing.

HPE initially launched the DL145 Gen11 line in 2024, which became the basis for its “Edge-in-a-box” solution it launched with Spectro Cloud. This latest iteration is basically a more compact version of the ProLiant DL385/380a servers unveiled last year.

There is also a new DL145 iteration tied to Microsoft Azure Local for users deploying Azure services to edge sites and that can take advantage of Azure Local Disconnected Operations.

HPE also revealed an “environmental ruggedization option kit” that bolsters the ProLiant Edge systems to operate in harsh outdoor environments. That ruggedization includes a temperature range from minus-40 degrees Celsius to 55 degrees Celsius, up to 95% humidity, and can survive “heavy vibration from aircraft or ground vehicles, environmental contaminants, and electromagnetic interference (EMI).”

The DL145 Gen11 platform and ruggedization kits are available immediately; the Azure Local iteration of the DL145 is set for availability next month; and The EL2000 chassis and EL220/EL240 servers are set for availability later this year.

They will also be available through HPE Financial Service’s “90/9 Advantage” program that offers no payments for 90 days and an additional 9 months at 1%.

Krista Satterthwaite, SVP and GM for compute at HPE, added that these platform updates also tap into the vendor’s networking division in supporting devices that can “connect in a distributed environment to deliver data securely, to be processed, to enable real value, like AI inferencing, analytics, and automation where it’s needed.”

CORRECTION: This story has been updated to correct the spelling of Krista Satterthwaite's last name.