Power servers

IBM turns POWER HPC momentum up to 11

Share this post:

It has been just one year since IBM reentered the high-performance computing (HPC) market with our new line of IBM POWER “LC” Linux servers powered by the POWER8 CPU.  I am happy to report that we already support hundreds of HPC applications, including seven of Intersect360’s Top 10 HPC applications. And we’ll have one more from their list to be added within this quarter.

Top 10 HPC Applications per Intersect360 Research report based on HPC User Site Census

Application Name Application Type Availability on POWER CPUs
Ansys Fluent Computational Fluid Dynamics Not available
Gaussian Computational Chemistry Q4 2016
GROMACS Computational Chemistry Now
NAMD Molecular Dynamics Now
Apache Hadoop Distributed Data Analytics Now
VASP Computational Chemistry Now
WRF Weather Forecasting Now
Simulia Abaqus Computational Mechanics Not available
OpenFOAM Computational Fluid Dynamics Now
AMBER Molecular Dynamics Now

Most of these applications are also accelerated by GPU accelerators, thereby enabling users to take advantage of the high-speed NVIDIA NVLink interface on the new IBM S822LC for HPC server. In addition to these, popular applications like the LS-Dyna finite element simulation application from LSTC and the Julia computing language for technical simulations are also available on POWER.

New benchmarks for HPC applications

IBM continues to benchmark HPC applications on the IBM S822LC for HPC server. As the chart below shows, we continue to see promising performance results with the new NVIDIA Tesla P100 Pascal GPU, along with the high-speed NVIDIA NVLink interface embedded in the new POWER8 CPU, on a range of scientific and technical computing applications.

power-hpc-benchmark-comparison

StoneRidge Technologies built the next-generation ECHELON oil and gas reservoir simulator from the ground-up to take advantage of GPU accelerators. The simulator is also available on the POWER8 server with NVIDIA NVLink.

Mass adoption of POWER8 servers by HPC clients

The low cost and wide availability of high-performance systems like the IBM S822LC for HPC server with NVIDIA NVLink technology helps researchers all over the world to deploy supercomputers. Recently, Yachay University in Ecuador deployed a POWER8 supercomputer for advancing research and scientific computing in South America. In addition, POWER8 HPC servers are being deployed widely by clients ranging from Oak Ridge National Labs, GENCI in France, SC3 Supercomputing Center in Turkey, cloud provider Nimbix, the University of Michigan, Lawrence Livermore National Labs and many more.

Get started today with the POWER8 NVLink Pascal Server today

An easy way to get started with the new IBM S822LC for HPC is to try one on the Nimbix POWER8 cloud. I look forward to hearing about your experiences with this server, so please post your comments below.

See you at Supercomputing 2016 in Salt Lake City!

 

Graph notes: Power System S822LC; 10 cores (2 x 10c chips) /  160 threads, POWER8 with NVLink; 2.86 GHz, 256 GB memory, 2 x 1TB SATA 7.2K rpm HDD, 2-port 10 GbEth, 2x Tesla P100 with NVLink GPUs;  Ubuntu 16.04. Competitive stack: 2x Xeon E5-2640 v4; 20 cores (2 x 10 chips) /  40 threads; Xeon E5-2640 v4;  2.4 GHz; 256 GB memory, 1 x 2TB SATA 7.2K rpm HDD, 2-port 10 GbEth, 2x Tesla K80 GPU, Ubuntu 16.04

Add Comment
No Comments

Leave a Reply

Your email address will not be published.Required fields are marked *

More Power servers Stories

Making deep learning easy and accessible to enterprises

As much as we talk about the power of deep learning, most companies are still warming up to the technology. There’s no denying the promise it offers, but realizing business value can be challenging. We’re about to change that, making it easier for leaders to harness insights with deep learning from the deluge of data. […]

Continue reading

Proposition: No speed limit on NVIDIA Volta with rise of AI

This is an era of data-centric computing.  There’s no doubt about that.  For those in hardware engineering who embraced disruption over the past few years – my colleagues at IBM Cognitive Systems and our partner NVIDIA included – this sudden rise of AI-inspired applications is a once or twice-in-a-career thrill. From a practical perspective, we […]

Continue reading

Five important considerations for HPC deployments

Architecting superior high-performance computing (HPC) clusters requires a holistic approach that responds to performance at every level of the deployment. HPC workloads put massive computational demands on your infrastructure: they involve analyzing huge data sets to solve complex problems, and that means they require supercomputing capabilities. In short, there are many facets to the HPC […]

Continue reading