ACADEMIA
Nanyang Technological University deploys cluster powered by IBM
Nanyang Technological University (NTU) February 11 opens its much-anticipated High Performance Computing (HPC) Centre to support the university's growing international research profile and capacity, especially in the area of sustainability.
The HPC Centre, which is based on the first IBM System x iDataplex cluster in ASEAN and powered by the Intel Xeon processor 5500 series, reduces electricity consumption as it can automatically adjust to specified energy usage levels and specified transaction speeds. NTU's supercomputer is ranked the 6th most energy-efficient in the world based on x86 architecture -- the universal platform found in computers today, and the 29th most energy-efficient system on the Green500 list at 274.64 Mflops (millions of floating point operations per second) per watt.
Unlike typical data centres which require a lot of energy to cool the operating environment, NTU's HPC system maximises performance with a unique water-cooled technology -- IBM's Rear Door Heat eXchanger for the iDataplex Rack. This eliminates the need for computer-room air conditioners, allowing for room-temperature operation. The result is a reduction in electrical consumption by more than 30 per cent when compared with the standard precision cooling system found in most supercomputers.
With one of the world's fastest supercomputers with a measured computing power (Rpeak) of over 29 teraflops (trillion mathematical calculations per second), NTU is now exploring more possibilities in leading-edge research and innovation. The research topics include developing future energy sources, studying global climate change, designing new materials, and understanding biological systems and the physics of complex socio-economic systems, among others. More can also be achieved in research such as in the modelling of volcanic activities, understanding the earth's tectonic movements and water treatment processes, as well as the simulation of flight dynamics.
"Prior to installing the supercomputer, pockets of computing capabilities were located in Schools on campus which compete for space and financial resources. The establishment of the supercomputer brings under one roof a centralised large-scale computing facility to the 2,800-strong research community on campus. It will also pave the way for a wider range of complex multidisciplinary research endeavours and more opportunities for research interaction within as well as outside of NTU," says NTU Provost Professor Bertil Andersson. NTU's green implementation has received much interest from local industries and has spurred the interest of organisations such as DSO, A*STAR and Singtel to set up a similar green data centre.
Sharing how the system has benefited his research, Assistant Professor Mu Yuguang from NTU's School of Biological Sciences says, "During the pilot phase we extensively tested the supercomputer's performance. We found two prominent features to be very attractive: the speed of each single central processing unit (CPU) and efficient parallelisation capacity, which means that we can study larger and longer processes required in our research and reduce the time needed to collect and analyse the data."
The HPC Centre, which is based on the first IBM System x iDataplex cluster in ASEAN and powered by the Intel Xeon processor 5500 series, reduces electricity consumption as it can automatically adjust to specified energy usage levels and specified transaction speeds. NTU's supercomputer is ranked the 6th most energy-efficient in the world based on x86 architecture -- the universal platform found in computers today, and the 29th most energy-efficient system on the Green500 list at 274.64 Mflops (millions of floating point operations per second) per watt.
Unlike typical data centres which require a lot of energy to cool the operating environment, NTU's HPC system maximises performance with a unique water-cooled technology -- IBM's Rear Door Heat eXchanger for the iDataplex Rack. This eliminates the need for computer-room air conditioners, allowing for room-temperature operation. The result is a reduction in electrical consumption by more than 30 per cent when compared with the standard precision cooling system found in most supercomputers.
With one of the world's fastest supercomputers with a measured computing power (Rpeak) of over 29 teraflops (trillion mathematical calculations per second), NTU is now exploring more possibilities in leading-edge research and innovation. The research topics include developing future energy sources, studying global climate change, designing new materials, and understanding biological systems and the physics of complex socio-economic systems, among others. More can also be achieved in research such as in the modelling of volcanic activities, understanding the earth's tectonic movements and water treatment processes, as well as the simulation of flight dynamics.
"Prior to installing the supercomputer, pockets of computing capabilities were located in Schools on campus which compete for space and financial resources. The establishment of the supercomputer brings under one roof a centralised large-scale computing facility to the 2,800-strong research community on campus. It will also pave the way for a wider range of complex multidisciplinary research endeavours and more opportunities for research interaction within as well as outside of NTU," says NTU Provost Professor Bertil Andersson. NTU's green implementation has received much interest from local industries and has spurred the interest of organisations such as DSO, A*STAR and Singtel to set up a similar green data centre.
Sharing how the system has benefited his research, Assistant Professor Mu Yuguang from NTU's School of Biological Sciences says, "During the pilot phase we extensively tested the supercomputer's performance. We found two prominent features to be very attractive: the speed of each single central processing unit (CPU) and efficient parallelisation capacity, which means that we can study larger and longer processes required in our research and reduce the time needed to collect and analyse the data."