Acer completes fastest supercomputer in Taiwan Ranks #42 in Top500® - 177 TFLOPS supercomputer, scalable with flexibleservices
Acer Inc. has completed construction of the 177 TFLOPS ALPS (Advanced Large Parallel System) supercomputing system for Taiwan’s National Center for High-performance Computing (NCHC). The system is the fastest supercomputer in Taiwan, and will be used to support research in science and engineering, including the fields of chemistry, disaster simulation, biotechnology, alternative energy, life sciences and more.
NCHC is the leading research institute in Taiwan for HPC. The center works with academic and enterprise institutes to take part in joint research and HPC service support. To further their service and support for the growing HPC research needs, the center sought to build a new supercomputing system that can offer flexible HPC resources for the research needs of Taiwan. The contract was awarded to Acer in November 2010.
Acer Corporate President, Jim Wong, emphasizes that “this system was a great opportunity for Acer to demonstrate our computing ability in the most demanding environments. Supercomputing is a benchmark for all fields of IT and proves Acer servers are ready to carve their space in datacenter environments.”
Regarding the build, NCHC’s director, Kuo-Ning Chiang, explained: “The decision to build this machine was made out of necessity. NCHC provides supercomputing resources to Taiwan’s many research groups and universities, and our previous machine with only 20 TFLOPS had a long waiting list for running computing jobs. The new system by Acer not only satisfies the current demand, but gives us a lot of room to grow.”
NCHC’s deputy director, Weichang Huang, was closely involved in the build and oversaw much of the work. He comments: “We greatly appreciate having Acer and its partners’ experience and expertise on this difficult case. Under tight deadlines, they were able to work with NCHC teams to complete this complex system. In addition to the largest system ever built in Taiwan, the experience learned from this effort has offered a great deal of knowledge for both NCHC and Acer, boosting the expertise of the HPC community of Taiwan to the next level.”
“Acer’s strong portfolio of High Performance and Technical Computing systems, integrated with industry leading technology partners, allowed us to deploy a custom HPC infrastructure able to support a wide range of applications to meet the needs of a diverse user community,” said Gianluca Degliesposti, Acer Vice President of Global Server Business Development.
The ALPS supercomputing system, nicknamed Windrider, was constructed through the consolidated efforts of Acer, AMD, Qlogic, DataDirect Networks, and Platform Computing.
The Cluster system uses the latest AMD Opteron™ 6000 Series processors and has a total of 8 compute clusters, 1 large memory cluster, and over 25,000 compute cores.
In order to meet a broad set of HPC needs, the system’s architecture is flexible to grow or shrink depending on the computing demand, and since all requests are off loaded from clients located all over Taiwan and the surrounding Asia region, the system advances the concept of HPC in the cloud.
The ALPS supercomputing system is completing user acceptance testing, and will begin fulfilling Taiwan’s research needs in July 2011. It is currently listed as number 42 in the June 2011 update to the Top500® List.