In a collaborative project with Department of Computer Science, the first phase of the project is implemented by early 2010. A new 1920 cores (based on Intel Nahalem 2.53 CPU) of 64-bit Linux Grid point system is installed to support System Research Group (SRG) of Department of Computer Science and Multiple Disciplines Research Project (MDRP) for HKU researchers, with a total of theoretical peak performance value of 19.43 Tera-FLOPS (and measured Rmax of 14.218 Tera-FLOPS). This system consists of latest hardware such as Intel Nehalem Quadcore CPU, 10Gb Ethernet switch and 4X DDR Infiniband communication. The second phase implementation is further implemented at late 2010, additional 336 cores based on Intel Westmere 2.66GHz CPU and 12 x NVIDIA Tesla M2050 GPUs together with 4X QDR Infiniband communication are integrated to the first phase system, making a total of theoretical peak performance value of 29.15 Tera-FLOPS for the whole Gridpoint system.
It is a UGC Special Equipment Grant project and is an extension of China National Grid (CNGrid) in Hong Kong. The project is initiated from the Computer Science Department, teaming up with the Information Technology Services, together with the support from a number of researchers from Department of Chemistry, Department of Statistics and Actuarial Science, etc.
Combining 2 phases of implementations, for the facility offering to the Multiple Disciplines Research Project (MDRP) by Information Technology Services, it consists of 144 blade servers organized in 10 chassis, making up a total of theoretical peak performance value of 12.402 Tera-FLOPS (and measured Rmax of 9.293 Tera-FLOPS). Each board is equipped with TWO 64-bit Nehalem Quadcore running at 2.53 GHz or TWO 64-bit Westmere 6-core running at 2.66GHz for supporting the research project for multiple departments. Its computational power is more than 6 times of the previous HPCPOWER2 64-bit cluster.
- System Configuration
- Software/Applications Available
- User Guide
- Usage Policy
- Inauguration Ceremony of HKU Grid Point for Systems Research and Applications in Multiple Disciplines