Skip to content

NVIDIA Introduces the DGX Spark HPC Developer Kit at SC26

NVIDIA BlogFrom the source· Allen Bourgoyne·· 2 hours agoPicksNVIDIA DGX Spark Developer Kit Turns a Desk Into a Supercomputing Lab
Original
Image: NVIDIA Blog
Who
NVIDIA
What
introduces the DGX Spark HPC Developer Kit, a four-node cluster for researchers
Quote
At SC26, an annual supercomputing conference taking place this week in Chicago, NVIDIA is introducing the DGX Spark HPC Developer Kit for researchers, academics and developers.

What happened

NVIDIA is introducing the DGX Spark HPC Developer Kit at SC26, a four-node cluster designed to run multi-node scientific workloads, AI models and agents at researchers’ desks. Four systems provide 512GB of combined memory and 20TB of NVMe storage, with direct ConnectX-7 links that require no network switch. The kit is on display at SC26 and available for sale; an NVIDIA HPC SDK installation playbook is coming soon.

Why it matters

Researchers with fixed compute allocations can run and debug multi-node scientific workloads locally, while keeping unpublished results and proprietary datasets on premises and reserving shared cluster time for production runs.

Read next

See the full storyNVIDIA Introduces the DGX Spark HPC Developer Kit at SC26

The story so far

NVIDIA introduced the DGX Spark HPC Developer Kit at the SC26 supercomputing conference in Chicago, making the four-node desktop cluster available for sale through top OEM partners and the NVIDIA Marketplace. Designed for researchers, academics, and developers running multi-node scientific workloads, AI models, and agents, the kit combines four systems with 512GB of memory and 20TB of NVMe storage. The nodes connect via direct ConnectX-7 links without requiring a network switch, NVIDIA said. The kit is on display at SC26, and NVIDIA said an installation playbook for building and running the NVIDIA HPC SDK, its scientific computing software development kit, is coming soon.

This story is in 11/17 AI DailyRead the whole issue