Watch video

Industries We Serve

Advanced HPC Solutions

Advanced HPC Products

Featured Products

Gemini™ High Availability Storage Solution

Gemini is a complete, next-gen high availability storage solution featuring a 1U dual controller RAID head engineered to seamlessly connect multiple JBODs with up to 512 drives.

Gemini ensures compatibility with the industry's leading parallel file systems -- including BeeGFS, IBM GPFS and Lustre – as well as superior levels of RAID performance and scalability; enterprise-level data protection and high density storage in a compact and efficient form factor.

As configured, Gemini includes:

  • 12Gb/s SAS-based 1U RAID head.
  • 1U 19-inch rackmount chassis.
  • HGST Ultrastar Data60® Hybrid Storage Platform.
  • Hot-swap and redundant components for increasing availability.
  • Support for multiple channel host options; up to 8GB cache per controller, optional FBM/BBM cache; and HDD firmware download.

Providing up to 720TB of raw storage, the HGST Ultrastar Data60 Hybrid Storage Platform used in Gemini is ideally suited for the demanding storage needs of enterprise customers that need dense, shared HDD or hybrid storage.

Gemini™ High Availability Storage Solution

Orion GPU-accelerated Storage Appliance

Advanced HPC, in partnership with Nyriad, posed a paradigm-shifting question: Could a GPU be used to significantly accelerate storage processing?
With the advent of Orion, the world’s first GPU-accelerated storage appliance, the answer is a resounding “Yes”. Supporting 100 HDDs in a 4U rack space, Orion features:

  • NVIDIA’s Tesla P4 GPU
  • 64GB of NVDIMMs
  • Thunder SX FA100 Chassis

Nyriad’s NSULATE software, at the heart of Orion, enables real-time hyperscale erasure coding up to 255 parity and real-time corruption recovery. Plus, since NSULATE is specifically designed for hyperscale storage, big data and high performance computing, Orion is ideally suited to tackle the most challenging storage processing demands faced by today’s leading-edge enterprise.

Orion GPU-accelerated Storage Appliance

NVIDIA Tesla V100

NVIDIA Tesla V100 is the world’s most advanced data center GPU ever built to accelerate AI, HPC, and Graphics.

  • 5120 CUDA cores
  • 640 New Tensor Cores
  • 7.5 TeraFLOPS double-precision performance with NVIDIA GPU Boost
  • 15 TeraFLOPS single-precision performance with NVIDIA GPU Boost
  • 120 TeraFLOPS mixed-precision deep learning performance with NVIDIA GPU Boost
  • 300 GB/s bi-directional interconnect bandwidth with NVIDIA NVLink
  • 900 GB/s memory bandwidth with CoWoS HBM2 Stacked Memory
  • 16 GB of CoWoS HBM2 Stacked Memory
  • 300 Watt
NVIDIA Tesla V100

Product Categories

News & Insights

View all