GPU

Rescale's Turnkey Cloud HPC Platform Now Offers NVIDIA Tesla V100 GPU With NVLink

Grazed from Rescale

Rescale, the global provider of HPC in the cloud, today announced that NVIDIA Tesla V100 GPU accelerators and NVIDIA NVLink high-speed interconnect technology are now available on Rescale's ScaleX turnkey cloud platform for AI and high performance computing (HPC). The new GPUs are hosted by Amazon Web Services (AWS) and as dedicated bare metal nodes by SkyScale, a Rescale partner. All resources are accessible on an hourly basis with Rescale's more than 200 pre-installed and pre-tuned HPC applications.

With the addition of the Tesla V100 to the ScaleX platform, Rescale users gain instant, hourly access to the fastest, most powerful GPU on the market. Based on the highly efficient NVIDIA Volta GPU architecture, the Tesla V100 is a big compute powerhouse, delivering 3x the training performance compared to its predecessor.

AWS Announces Availability of P3 Instances for Amazon EC2

Grazed from Amazon Web Services

Today, Amazon Web Services, Inc. (AWS), an Amazon.com company, announced P3 instances, the next generation of Amazon Elastic Compute Cloud (Amazon EC2) GPU instances designed for compute-intensive applications that require massive parallel floating point performance, including machine learning, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, genomics, and autonomous vehicle systems. The first instances to include NVIDIA Tesla V100 GPUs, P3 instances are the most powerful GPU instances available in the cloud. To get started with P3 instances, visit https://aws.amazon.com/ec2/instance-types/p3/.

P3 instances allow customers to build and deploy advanced applications with up to 14 times better performance than previous-generation Amazon EC2 GPU compute instances, and reduce training of machine learning applications from days to hours. With up to eight NVIDIA Tesla V100 GPUs, P3 instances provide up to one petaflop of mixed-precision, 125 teraflops of single-precision, and 62 teraflops of double-precision floating point performance, as well as a 300 GB/s second-generation NVIDIA NVLink interconnect that enables high-speed, low-latency GPU-to-GPU communication. P3 instances also feature up to 64 vCPUs based on custom Intel Xeon E5 (Broadwell) processors, 488 GB of DRAM, and 25 Gbps of dedicated aggregate network bandwidth using the Elastic Network Adapter (ENA).

AMD works with Amazon to create virtualized graphics in the cloud

Grazed from VentureBeat. Author: Dean Takahashi

Amazon Web Services has chosen to use Advanced Micro Devices’ graphics technology to run graphics software in the cloud. The net result is that it could become a lot cheaper to process graphics-intensive apps in the cloud, rather than on a local machine.

AMD said in a blog post that it’s no secret most enterprise applications — from standard Windows productivity apps to engineering software — run better when they are accelerated by graphics processing units (GPUs). Traditionally, those apps have run on heavy-duty workstations on local machines.
 

With Volta, NVIDIA Pushes Harder into the Cloud

Grazed from Top500. Author: Michael Feldman.

Amid all the fireworks around the Volta V100 processor at the GPU Technology Conference (GTC) last week, NVIDIA also devoted a good deal of time to their new cloud offering, the NVIDIA GPU Cloud (NGC). With NGC, along with its new Volta offerings, the company is now poised to play both ends of the cloud market: as a hardware provider and as a platform-as-a service provider.

At the heart of NGC is a set of deep learning software stacks that can sit atop NVIDIA GPUs – not just the new Tesla V100, but also the P100, or even the consumer-grade Titan Xp. The stack itself is comprised of popular deep learning frameworks (Caffe, Microsoft Cognitive Toolkit, TensorFlow, Theano and Torch), NVIDIA’s deep learning libraries (cuDNN, NCCL, cuBLAS, and TensorRT), the CUDA drivers, and the OS...

New Version of Univa Grid Engine 2x Faster Than Sun Grid Engine 6.2u5

Grazed from Univa

Univa, a leading innovator of workload management products, today announced the general availably of its Grid Engine 8.5.0 product. Univa’s enterprise customers continue to push the boundaries of scalability and performance and Univa has responded. With Univa Grid Engine 8.5.0 benchmarks show an impressive 2x performance improvement in scheduling and processing of all types of workload versus previous versions of Grid Engine. Enterprises will now see immediate results with less wait time and more work completed.

In addition to scalability and performance improvements, this release includes resource map enhancements for selecting host-based resources such as GPUs and improvements to Docker device selection, allowing GPUs to be mapped into Docker containers. Users can now place GPU applications into Docker containers and run them on any Docker-enabled host in the cluster.
 

Kinetica and Nimbix Team Up to Offer GPU Computing in the Cloud for Enterprise Customers

Grazed from Kinetica and Nimbix

Kineticaprovider of the fastest, in-memory database accelerated by GPUs, today announced its real-time analytics and visualization solution is immediately available on the Nimbix Cloud.  Providing instant results and visualized insights across massive streaming datasets, Kinetica on the Nimbix Cloud can be launched in seconds and is the ideal solution for GPU-accelerated analytics.

"Kinetica on the Nimbix Cloud harnesses the power of parallel GPUs to deliver real-time analytics and data written to Kinetica is automatically routed to parallel connections across the cluster," said Amit Vij, cofounder and CEO, Kinetica.  "The full Kinetica stack can be provisioned with a couple of mouse clicks from the Nimbix console or launched and automated with JARVICE's powerful task API."

Liquidware Labs Announces the Release of Stratusphere UX 5.8.6 - Now with vGPU Monitoring Powered by NVIDIA GRID

Grazed from Liquidware Labs

Liquidware Labs, a leader in desktop transformation solutions, today announced the general availability of Stratusphere UX 5.8.6 with a number of new features, including new virtual GPU (vGPU) metrics as well as dashboard enhancements, greater security and more. The company is demonstrating the software at Citrix Summit 2016 Anaheim this week (Booth #408).

"Stratusphere UX continues to lower the barrier of entry for the diagnostics and monitoring of end-user workloads" said David Bieneman, CEO, Liquidware Labs. "The solution's ability to focus on all users, machines and applications, while defining a quantitative metric for user experience, is the defining aspect of Stratusphere. And now, with more granular vGPU metrics and enhanced dashboard views, Stratusphere UX offers greater capabilities in support of trending, optimizing and diagnosing the challenges faced in next generation end-user workspaces."
 

Nimbix and Bitfusion Deliver Industry's Most Affordable High Performance GPU Resources in the Cloud

Grazed from Nimbix an Bitfusion

Nimbix and Bitfusion today announced the immediate availability of its new combined solution to offer more choices to application developers looking for high performance GPU accelerators on an on-demand basis. The Nimbix Cloud, powered by JARVICE, now integrates Bitfusion Boost to offer lower cost accelerator resources for developing compute hungry machine learning, analytics, and photorealistic rendering algorithms.

The solution uses Bitfusion Boost's software defined compute technology to enable higher horsepower virtual GPU instances with costs significantly lower than anywhere in the cloud. The solution will also permit the combining of interesting CPU-memory-GPU configurations in the future. A developer will be able to combine the largest memory machine with an unlimited number of GPU nodes together to form a single powerful instance capable of handling a much larger job than before.

Cloud Computing: AMD Unveils Virtualized GPUs, New Quiet Chips

Grazed from eWeek. Author: Jeffrey Burt.

Advanced Micro Devices is continuing a steady drumbeat of product releases as officials look to make the chip maker a larger player in everything from high-end gaming PCs and virtual desktops to cloud computing. The company on Feb. 1 unveiled virtualized graphics cards that can be housed in data center servers and delivered as needed to a range of devices, including remote workstations and other client systems. A day later, officials introduced two new desktop processors—the A10-7860K and Athlon X4 845—featuring AMD's new Wraith Cooler thermal system that they said generate less than a 10th of the noise of previous chips.

"Customers can build a capable, near-silent online gaming PC in a surprisingly small form factor for an unexpectedly low price," Kevin Lensing, corporate vice president and general manager of AMD's Client Business Unit, said in a statement. "The low-power requirements mean you can build a tiny system perfect for a small dorm room, for use in the living room as a home theatre PC … and portable enough to take to a LAN party."...

IBM unleashes NVIDIA GPU on SoftLayer cloud for faster processing power

Grazed from CloudTech.  Author: James Bourne.

While the rest of the tech media marvelled at the world’s first seven nanometre chips unveiled by IBM earlier this week, the Armonk giant has quietly pushed out a couple of impressive cloud announcements alongside it.  IBM – who, let us not forget, wants to be seen as a cloud-first organisation – has announced the availability of NVIDIA Tesla K80 dual-GPU accelerators on bare metal cloud servers. In other words, it provides supercomputing powers to the SoftLayer cloud without the need for companies to expand their existing infrastructure.
 
GPU, in a cloud environment, works alongside a server’s CPU to accelerate application and processing performance. Combining the two instead of using a CPU alone, utilising thousands of small efficient cores, results in faster processing of information...