Today NICE announced plans to demonstrate their upcoming Desktop Cloud Visualization (DCV) 2014 technology at the AWS re:Invent conference as well as SC14.
Leveraging special-purpose hardware-accelerated stream processing from NVIDIA GRID cards, the solution provides drastically reduced bandwidth usage for high-end graphics. DCV 2014 offers 3 to 10 times bandwidth savings over DCV 2013 and improved network latency tolerance, thus expanding the market for 3D visualization in the cloud to entirely new use cases.
NICE is the first company to leverage the new NVIDIA Kepler-specific H.264 encoder both on Linux and Windows,” said Andrew Cresci, General Manager, NVIDIA GRID. “Our collaboration with NICE is enabling new users and businesses to leverage the NVIDIA GRID technologies offered in the Amazon Public Cloud.”
DCV allows engineers and scientists to take full advantage of the high-end graphics cards, fast I/O performance, and large memory nodes hosted in public or private 3D clouds such as Amazon Web Services (AWS), rather than having to wait for their workstations to be upgraded with the latest hardware. Coupled with NICE’s EnginFrame, DCV allows ISVs and customers to build end-to-end, interactive and batch, HPC or design processes in the cloud. NICE DCV also provides desktop sharing capabilities — concurrent access to the same 3D interactive session among multiple users — that enable distributed teams to collaborate in the cloud.
Amazon Elastic Compute Cloud (Amazon EC2) G2 instances, which leverage the NVIDIA GRID GPUs based on NVIDIA Kepler architecture, provide high-end graphics capabilities for both Windows and Linux applications that are accessible from any network and client. If memory size is an issue, DCV can combine a G2 instance with a Windows R3 instance to provide large memory and graphics acceleration. Being able to visualize Windows applications in the AWS Cloud enables engineers to quickly and easily start or expand CAD/CAE projects, while keeping infrastructure costs down.
DCV uses H.264 compression and leverages the Kepler architecture of NVIDIA GRID in order to provide significant acceleration for image encoding. GPU-accelerated compression further allows high frame rates, more accuracy, and reduction in overall bandwidth usage of the visualization stream, to the extent that 3D visualization is now possible over low bandwidth wide-area links.
The NICE DCV 2014 will be released in Q4 of 2014.
Sign up for our insideHPC Newsletter.
