Product Announcements and
Upcoming Events
|
|
|
|
NVIDIA® Tesla® P100 GPU Accelerator
|
|
The NVIDIA Tesla P100 GPU Accelerator is the most advanced GPU to date. Featuring the breakthrough NVIDIA Pascal architecture, the Telsa P100 is the most flexible and versatile GPU NVIDIA have ever released. It is available in the traditional PCIe form factor as well as their new socketed SXM2 form factor (pictured above). Additional features include:
- Exceptional performance with up to 5.3 TFLOPS double- and 10.6 TFLOPS single-precision floating-point performance.
- NVLink enables a 5X increase in bandwidth between Tesla Pascal GPUs and from GPUs to supported system CPUs (compared with PCI-E). Up to 80GB/s (160GB/s bi-directional).
- High-bandwidth HBM2 memory provides a 3X improvement in memory performance compared to Kepler and Maxwell GPUs.
- Pascal Unified Memory allows GPU applications to directly access the memory of all GPUs, as well as all of system memory (up to 512TB).
|
|
|
The variety of new products incorporating NVIDIA Tesla P100 GPUs may be overwhelming. We've drawn up diagrams and broken down the details below. Contact us to discuss the best option for your needs.
|
|
|
NumberSmasher 1U Tesla GPU Server with NVLink
|
|
- With 2 or 4 NVIDIA Tesla P100 GPUs with 80GB/s NVIDIA NVLink®
- Additional slots for EDR/FDR InfiniBand, 10G/40G Ethernet, and high-speed storage
- NVIDIA CUDA® Toolkit installed and configured – ready to run GPU jobs!
- Up to 44 Intel processor cores with Intel Xeon E5-2600v3/v4 CPUs
- Up to 1 TB system memory
|
|
|
OpenPOWER GPU Server with IBM POWER8 + NVLink
|
|
- Two or four top-performance NVIDIA Tesla P100 GPUs with NVLink
- 80GB/s NVLink connectivity for data-intensive and multi-GPU applications
- NVIDIA CUDA Toolkit installed and configured – ready to run GPU jobs!
- Up to 20 IBM POWER8 processor cores (each supporting 8 threads)
- Up to 1TB system memory with a total throughput of 230GB/s
- Support for high-speed InfiniBand fabrics and ethernet connectivity
|
|
|
NVIDIA DGX-1 Deep Learning System
8 NVIDIA Tesla P100 GPU Accelerators with NVLink
|
|
- Deep Learning libraries and frameworks built-in
- Software stack includes:
- DIGITS training system (version 3 or newer)
- NVIDIA Deep Learning SDK with CUDA 8.0 and cuDNN v5
- Cloud management software/services:
- NVIDIA Cluster Portal (cloud or onsite)
- Online application repository with the major deep learning frameworks
- NVDocker containerized app deployment
- Managed app container creation and deployment
- Multi-Node management with telemetry, monitoring and alerts
|
|
|
NVIDIA Tesla PCIe GPU Servers
|
|
Our wide variety of servers accommodate any need or any budget. All Microway NVIDIA GPU servers are configured with:
- NVIDIA Tesla P40, K80, and P100 GPU Accelerators*
- Up to 44 processor cores with Intel E5-2600v3/v4 CPUs
- Up to 1.5TB memory
- NVIDIA CUDA SDK installed and configured – Ready to run CUDA jobs!
*For additional GPU options, please inquire.
|
|
Octoputer 4U Server with 8 GPUs
- Up to 8 NVIDIA Tesla P100, P40, or K80 GPUs
- Additional slots for EDR/FDR InfiniBand, 10G/40G Ethernet and storage controllers
- Up to 128GB GPU Memory
- Up to 1.50 TB memory
- Up to 96 TB storage built-in
|
|
|
- Up to 10 NVIDIA Tesla P100, P40, or K80 GPUs
- Additional slots for EDR/FDR InfiniBand, 10G/40G Ethernet, or high-speed storage
- Up to 160GB GPU Memory
- Up to 1.50 TB memory
- Up to 96 TB storage built-in
|
|
|
|
Test Drive the New NVIDIA Tesla P100!
|
|
At Microway, we pride ourselves on always making state-of-the-art technology available to our customers. Our GPU Test Drive Cluster now has the latest NVIDIA Tesla GPUs so that you can benchmark your code to get an idea of exactly how much faster it will run!
Schedule your test drive today! Submit a request online at microway.com/gpu-test-drive.
|
|
|
Upcoming Events Where You Can Meet Us
|
|
|
The NVIDIA GPU Technology Conference (GTC) in Washington, DC is the premier Deep Learning event offering a focus on GPU computing and its impact in areas such as high-performance computing, autonomous driving, and artificial intelligence. This two day event will offer five focused tracks and 50+ AI sessions.
|
|
|
|
COMSOL Conference - Boston 2016
|
|
|
This annual event brings together experts from around the globe in Multiphysics Modeling & Simulation. Connect with the makers of COMSOL Multiphysics and participate in over 50 training opportunities and events explicitly designed for the engineering, CFD, FEA, and modeling community.
|
|
|
|
Supercomputing 2016 - Salt Lake City
|
|
|
Sponsored by the ACM (Association for Computing Machinery) and the IEEE Computer Society, SC16 offers a complete technical program, courses and meetings for students and educators in HPC, and an exhibition that showcases the many ways high performance computing, networking, storage and analysis lead to advances in scientific discovery, research, education and commerce.
Visit us at Booth 2115
|
|
|
Please follow the links we've provided for the most detailed information on exciting new products and events in the HPC world.
As always, if you have any questions, Microway's experts are happy to offer advice and share our technical expertise. Feel free to call us when you design your next HPC cluster or WhisperStation.
|
|
|
|
|
|
