Fastest Compute Engines in the World

Up to 16 P100 Tesla GPUs in a single System

4 GPU SERVICE

Based on the One Stop Systems OSS-PASCAL4

The 1U GPU accelerated server (OSS-PASCAL4) sets a new standard for HPC compute power in a space efficient design. Supporting four of the latest Pascal-based NVIDIA P100 SXM2 GPUs, the OSS-PASCAL4 provides 21.2 TeraFLOPS of double precision performance for the most demanding HPC applications. For state-of-the-art deep learning workloads the OSS-PASCAL4 provides 84.8 TeraFLOPS of half precision performance. The GPU accelerated server also includes dual high-performance “Broadwell” E5 2690v4 2.6GHZ processors and a base configuration of 512GB of DDR4 memory scalable to 1TB. Four PCIe Gen3 slots are available for additional expansion and for scale out clusters can be provided using Infiniband networking.

8GPU SERVICE

Based on the One Stop Systems OSS-PASCAL8

The OSSPASCAL8 supports eight of the latest Pascal-based NVIDIA GPUs, providing 42.4 TeraFLOPS of double precision performance for the most demanding HPC applications. For state-of-the-art deep learning workloads the OSS-PASCAL8 provides 169.6 TeraFLOPS of half precision performance. The GPUs are interconnected with 80GB/s NVLink providing unsurpassed peer-to-peer bandwidth. The GPU accelerated server also includes dual high-performance “Broadwell” E5 2667v4 3.2GHZ processors and a base configuration of 512GB of DDR4 memory scalable to 2TB. Six PCIe Gen3 slots are available for additional expansion and for scale out clustering using Infiniband networking. The appliance includes one 2.5” removable 1.6TB NVMe SSD drives.

16 GPU SERVICE

Based on the One Stop Systems CA16000 and EOS Server

The CA16000 Compute Accelerator with sixteen NVIDIA® Tesla® GPU accelerators is employed in a variety of HPC applications including oil and gas exploration and financial services. It provides over 57,300 cores and 299.2 TFlops of half-precision compute power. The CA16000 occupies only 3U of rack space and connects directly to its host server through the latest technology PCIe x16 Gen3 connections. Four removable canisters house up to four full-height, full-length, PCIe x16 double-wide GPUs and one half-length, half-height IO card each. The system is powered by three 3000-watt redundant power supplies and includes an IPMI-based system monitor. The EOS One Stop Systems 2U server provides the high end Intel Xeon CPUs including dual “Broadwell” E5 2667v4 3.2GHZ processors and a base configuration of 512GB of DDR4 memory.

SKYSCALE SECURITY, RELIABILITY AND COMPLIANCE

SkyScale’s HPC systems are hosted in high-end, high-security, and highly reliable datacenters located in San Diego, CA and Houston, TX. The datacenters are audited annually to the SSAE-16/ISAE SOC1 Type 2, and SOC2 Type 2 standards, and they meet or exceed HIPPAA, PCI-DSS, and GLB compliance standards. SkyScale can provide infrastructure-related control documentation on request.

INDUSTRY SOLUTIONS

Deep Learning and A.I.

Financial

Oil and Gas

Life Science

Research

HPC Cloud Services With the Leadership Features You Need

Dedicated Hardware

You control your compute power with exclusive access to the entire platform. All the GPUs drive your applications. No virtualization, no sharing.

Secure

Housed in world class secure data centers with the highest security certification levels. Your equipment is always available and your data is always protected and secure. Download our Security and Reliability Document. 

Access to Experts

When you lease from SkyScale you also have access to the experts who understand your needs and can get you going quickly so you can focus on the work that truly drives the value of your business.

Save Time

Time to answer, time to do more, time to dive deeper, handle more data and predict more. Running your application on the fastest platforms minimizes your time to value.  Your engineers and scientists will be happier and more productive.

Save Investment

Only lease what you need when you need it. Save the capital investment and don’t worry about having idle time when you are between projects.  Save the cost of maintaining and staffing your own data center.

Highest Performance

With exclusive access to as many as 16 NVIDIA P100s in a single node you are running your applications on the world’s absolute highest performing multi-GPU platform in the world. Couple that with state-of-the-art high speed low latency NVMe storage and your project time just got a lot shorter.