Today IBM announced launched a new LC series of servers that infuse technologies from members of the OpenPOWER Foundation and are part of IBM’s Power Systems portfolio of servers. According to IBM, the new LC systems perform data analytics workloads faster and cheaper than comparable x86-based servers.
“Although the use of GPUs has generalized nowadays, including GPUs in current HPC clusters presents several drawbacks mainly related with increased costs. In this talk we present how the use of remote GPU virtualization may overcome these drawbacks while noticeably increasing the overall cluster throughput. The talk presents real throughput measurements by making use of the rCUDA remote GPU virtualization middleware.”
For some applications, cloud based clusters may be limited due to communication and/or storage latency and speeds. With GPUs, however, these issue are not present because application running on cloud GPUs perform exactly the same as those in your local cluster — unless the application span multiple nodes and are sensitive to MPI speeds. For those GPU applications that can work well in the cloud environment, a remote cloud may be an attractive option for both production and feasibility studies.
“Our vision is to deliver accelerated graphics and high performance computing to any connected device, regardless of location,” said Jen-Hsun Huang, co-founder and CEO of NVIDIA. “We are excited to collaborate with Microsoft Azure to give engineers, designers, content creators, researchers and other professionals the ability to visualize complex, data-intensive designs accurately from anywhere.”
As an open source tool designed to navigate large amounts of data, Hadoop continues to find new uses in HPC. Managing a Hadoop cluster is different than managing an HPC cluster, however. It requires mastering some new concepts, but the hardware is basically the same and many Hadoop clusters now include GPUs to facilitate deep learning.
Today IPSJ, Japan’s largest IT society honored Bill Dally from Nvidia with the Funai Achievement Award for his extraordinary achievements in the field of computer science and education. “Dally is the first non-Japanese scientist to receive the award since the first two awards were given out in 2002 to Alan Kay (a pioneer in personal computing) and in 2003 to Marvin Minsky (a pioneer in artificial intelligence).”
In a perfect world, there would be one version of all compilers, libraries, and profilers. To make things even easier, hardware would never change. However, technology marches forward, and such a world does not exist. Software tool features are updated, bugs are fixed, and performance is increased. Developers need these improvements but at the same time must manage these differences.
HPC developers want to write code and create new applications. The advanced nature of HPC often requires that this process be associated with specific hardware and software environment present on a given HPC resource. Developers want to extract the maximum performance from HPC hardware and at the same time not get mired down in the complexities of software tool chains and dependencies.
HPC and Beer have always had a certain affinity ever since the days when Cray Research would include a case of Leinenkugel’s with every supercomputer. Now, Brian Caulfield from Nvidia writes that a Pennsylvania startup is using GPUs and Deep Learning technologies to enable brewers to make better beer.