By Michael O’Neill (NVIDIA), Brandon Harrell (IBM), Deana Coble (IBM)
Traveling gives you plenty of time to think, it is usually a mix of thinking about work and missing home (such is the reality of life). In assembling this week’s blog for our series, our team had the challenge of describing for you the realities of visual computing with NVIDIA GRID and IBM NeXtScale, at the same time as traveling. However, the travels lead us to a great metaphor for answering that challenge.
One of the members of our team was thinking of his family back at home. Michael spoke of his “daughters’ bright eyes,” and mentioned the “subtle, soft tones in the colors”, of his better half’s hair. We all realized that it was these kinds of details that made a difference. The small visual cues are what makes things “real”, and makes things alive. It was also the key to helping us describe visual computing!
Visual Computing is a model that provides for the evolution of the datacenter, personal devices, and the user experience for a brilliant future. In fact, NVIDIA invented that model. NVIDIA provides a robust set of technologies to enable “realness” in computing. “Realness” that allows designers, engineers, and geologists to mold data into a car… a rocket ship… or to see below the Earth’s crust to find our next source of energy. The subtle details in these endeavors are paramount.
The modern tools of these trades are applications that leverage parallel computing. They require many processes and results to be delivered simultaneously, a task for which NVIDIA GPUs are inherently built and have been achieving for many years now. For example, with applications like these: PTC Creo, AutoDesk, Siemens NX, SolidWorks, and many more. The components pictured below provide a glimpse of the difference between a flat non-GPU delivery and the “realness” of GPU acceleration for these applications.
Left: Underachieving CPU-only details; Right: Extreme GPU-enabled details
Now imagine the components in Figure 1 are not a bicycle sprocket but rather a hydraulic component for an airplane or a new braking system for a new mini-van. The details and realness of what the engineer sees can be the difference between success and failure. This is one reason why NVIDIA GPU enabled visual computing is a requirement for business today and into the future.
NVIDIA Quadro has long been the industry standard for professional visual computing on a workstation. IBM NeXtScale and NVIDIA GRID bring this visual computing world to the datacenter. IBM NeXtScale allows you to leverage all of the agility that comes with virtualization (Server, Application, or VDI) on a rock solid server foundation and provides the power, GPU and CPU densities to keep costs in line. Imagine 6 compute nodes and 12 high performance NVIDIA GPUs in a mere 6U of rack space. Now stop imagining because it’s reality!
The NeXtScale n1200 chassis can house up to 6 Compute Nodes, each with a two-slot PCIe Native Expansion Tray attached,
allowing up to 12 GPU cards in just 6U of rack space. Pictured on right: NVIDIA GRID GPU options available.
This reality is made easily accessible through the NeXtScale System offering that includes features like these:
- Up to two of any of the 5 supported high-performance NVIDIA GPUs, including the GRID K1, K2 and Tesla K40
- Two standard 1Gb Ethernet ports, plus support for 2 additional adapters (such as 10Gb Ethernet or InfiniBand)
- Support for 2 of the latest Intel Xeon E5-2600 v2 processors and up to 256 GB of memory
- Integrated Management Module II (IMM2) to monitor server availability (including advanced predictive failure and health monitoring). Optional remote management feature for enhanced serviceability
Whether you are local or remote, traveling or at home, we hope you are enjoying the series about GPU accelerated data centers. There is still more to come, over the next two weeks, for these topics and what it means for you:
- GPU Accelerated Compute for Analytics (for Financial, Rendering, and so on)
- GPU Accelerated Big Data
NVIDIA GPUs and IBM NeXtScale Systems encourage you to keep it real – enjoy the visuals around you wherever you are! See you next week.
Use this hyperlink to subscribe to breaking news and newsletters about NVIDIA product updates and more, and use the following hyperlinks to see a virtual tour, product guide, product demo, and more information about IBM NeXtScale.
Michael O’Neill is an established leader for NVIDIA and provides specialized technical guidance to customers on NVIDIA GRID and Tesla GPUs in virtualized environments. He works closely with IBM to develop innovative solutions for graphical and compute heavy workloads.
Brandon Harrell is an IBM Technical Solutions Leader. He specializes in x86 Desktop and Application Virtualization as part of the Advanced Technical Sales team. Follow Brandon on Twitter at @BrandonEHarrell.
Deana Coble is an IBM Redbooks Technical Writer. She has worked over 20 years in the field of technology and collaborates with IBM subject matter experts to create IBM Redbooks publications on a variety of systems. Follow Deana on Twitter at @DeanaCoble.