One Stop Systems, Inc.

• SeriesOSS-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 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. GPU management and monitoring and software are preinstalled on the OSS-PASCAL4. The GPU accelerated server also includes dual high-performance “Broadwell” E5 2698v4 2.2GHZ 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 creating GPUltima clusters using IB or high speed Ethernet networking. The appliance also includes two 2.5” removable 1.9TB SATA SSD drives. HARDWARE HIGHLIGHTS 1U Chassis Dual Intel Xeon 3.2GHz CPUs Up to 1TB DDR4 LRDIMM System Memory Two 2.5” 1.9TB SATA SSDs Four Pascal GPU SXM2 with 80GBs NVLink Three x16 PCIe 3.0 slots One x8 PCIe 3.0 slot Two 2000W Titanium Power Supplies SOFTWARE HIGHLIGHTS Choice of Machine Learning Framework Caffe Torch Tensorflow Theano MLPython ML Dependencies (400MB Python) cuDNN (5.0 & 5.1) DIGITS Caffe on Spark CUDA & NVIDIA driver CUB (CUDA building blocks) NCCL GPU Management from Bright Computing Health Management Workload Integration

• SeriesCA2000
• Peak Double Precision Performance.2 TeraFLOPS 3.8 TeraFLOPS 1.43 TeraFLOPS 2.91 TeraFLOPS 4.7 TeraFLOPS
• Power Consumption250 W 300 W 235 W 300 W 250 W

The CA2000 Compute Accelerator with one or two NVIDIA® Tesla® GPU accelerators is employed in a variety of HPC applications including oil and gas exploration and financial services. Completely integrated with the GPUs most suited for a specific application, it's easy installation and tested reliability makes it superior to alternative products. The CA2000 occupies only 1U of rack space and connects to the host server through the latest technology PCIe x16 Gen3 connection. PRODUCT HIGHLIGHTS 1U High One Rear Panel PCIe x16 Gen3 Interface Remote System Monitoring Capability; monitor fans, temperature and voltages 1620 Watt Power Supply Superior Cooling with Eight 30 CFM Fans Choice of 1 or 2 NVIDIA Tesla GPUs

The GPUltima is a high-density, fully integrated computer cluster that is purpose-built for high performance computing (HPC) applications like financial trading, deep learning or machine learning, oil and gas exploration, virtual desktop infrastructure (VDI), defense and security, and academia. Where conventional computer cluster systems use CPUs as the primary data processor, the GPUltima employs numbers of GPU cards, providing 10 times the performance by adding thousands more cores. In addition, the GPUltima consumes about 90% less power than conventional systems and occupies 95% less rack space. The GPUltima petaflop edition requires only a single 42U rack and requires only 56kW of power. The GPUltima is completely 'application-ready', so that all the customer has to do is to add his application software to the servers and the system is ready to begin processing. The unique cluster management and monitoring software and the service and support packages that accompany the GPUltima make this a user-friendly system that allows the customer to begin his work without having to configure the cluster. The GPUltima is comprised of 'compute nodes'. Each compute node contains sixteen GPU cards and one or two dual socket servers, each with dual 'Broadwell' processors, producing 139 Teraflops of performance using NVIDIA Tesla K80s. The sixteen GPU cards communicate with each other through Infiniband EDR 100Gbs connections through a 1U Infiniband switch. The server(s) are cabled to the GPU enclosure through 128Gbs PCIe. The server communicates to the internet through Ethernet. Additional nodes can be added as needed. For more than one server an Ethernet switch is added for external communication. The clustering software allows complete manageability of the nodes as well as the individual GPU cards. The GPU monitoring and management software provides a 'single-pane-of-glass' management of the hardware, the operating system, HPC software, and users. With the Cluster Manager, system administrators can quickly get clusters up and running and keep them running reliably throughout their lifecycle — all with the ease and elegance of a full-featured, enterprise-grade cluster manager.