November 24, 2006
Synopsys Inc., a provider of semiconductor design software, has announced it is the first electronic design automation (EDA) company to be included on a list of sites operating the world's most powerful supercomputers. The Synopsys-built supercomputer was ranked as the 242nd most powerful supercomputer in the world based on the Linpack benchmark.
The Synopsys machine was built to perform production runs with intense compute needs. Synopsys constructed the supercomputer in less than four months using commodity Linux servers, delivering up to ten times the software performance of traditional legacy computing systems in certain applications. Synopsys' supercomputer achieved benchmark results surpassing 3.7 teraflops -- roughly equivalent to 18,000 personal computers all working together at the same time to solve the same problem.
"Synopsys IT is focused on optimizing our existing computing infrastructure to enable our engineers to perform their jobs more effectively," said Debra Martucci, CIO and vice president of IT for Synopsys. "This supercomputing milestone illustrates how we are able to do more with the IT equipment we already have. By re-configuring our existing commodity hardware, we're exceeding our performance benchmarks and achieving greater than 50 percent global utilization rates for our server infrastructure."
Rather than purchase expensive equipment that would remain idle much of the year, the Synopsys IT team re-architected and re-deployed existing off-the-shelf hardware on a nightly basis to operate, baseline and tune the supercomputer. According to the company, this strategy maximized total cost of ownership while minimizing any impact to daily business operations. During more than one thousand scheduled system tests, the IT team constructed the testing requirements and measured and tuned the various configuration options. This allowed them to maximize operational performance with minimal additions to the network and compute infrastructure.
The Synopsys supercomputer, completed in September 2006, is comprised of 329 Linux servers connected by non-blocking standard Gigabit Ethernet. It was assembled from six pre-existing clusters of Linux servers used by Synopsys engineers during business hours.
In recent years the cost of supercomputing has plummeted, largely due to the rise of Linux and dramatic improvements in the performance of commodity computing servers, analysts say. Thanks to these price-to-performance gains, sales of high-performance computers grew at 24 percent to $9.2 billion in 2005, according to market research firm IDC (IDC, "Looking Back and Forward - What Has Changed in the Technical Computing Market?" Doc #204099, Nov 2006).
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?