Japanese Government Procurement
National University Corporation - Notice of request for submission of materialsUnified Memory Supercomputer System 1 set
This procurement is covered by the WTO Agreement on Government Procurement, Japan-EU Economic Partnership Agreement or Japan-UK Comprehensive Economic Partnership Agreement.
| Publishing date | Apr 08, 2024 |
|---|---|
| Type of notice | Notice of request for submission of materials |
| Procurement entity | National University Corporation - Ibaraki |
| Classification |
0014 Office Machines & Automatic Data Processing Equipment |
| Summay of notice | ⑴ Classification of the products to be procured : 14 ⑵ Nature and quantity of the products to be purchased : Unified Memory Supercomputer System 1 set ⑶ Type of the procurement : Rent ⑷ Basic requirements of the procurement : ・A Unified Memory Supercomputer System consists of parallel computing servers, a parallel file system, login servers, and other hardware and software for stable and efficient operation of the whole system. A Parallel computing servers must satisfy the following conditions. ⒜ Parallel computing servers consist of identical compute nodes. ⒝ A compute node consists of CPU, GPU, memory, NVMe SSD and other peripherals. ⒞ Theoretical peak performance of GPU must be equal to or higher than 60 TFLOPS in double precision floating point arithmetic. ⒟ CPU and GPU can cache-coherently access all memory in a compute node. Total memory capacity must be 128 GiByte or larger, and the maximum theoretical peak bandwidth must be 4 TByte/sec or higher. ⒠ Storage capacity of NVMe SSD should be 3.8 TByte or larger. The theoretical peak bandwidth must be 8 GByte/sec or higher. ⒡ Each compute node must be connected with a full bisection network interconnect, each link speed must be 400 Gbps or higher per GPU. ⒢ The total peak performance of double precision floating point must be 10 PFLOPS or higher. B A parallel file system must satisfy the following conditions. ⒜ The parallel file system must be connected to the interconnection network of all compute nodes. ⒝ The capacity of user available space must be 1 PByte or larger. ⒞ The theoretical peak bandwidth must be 100 GByte/sec or higher. ⒟ The maximum number of inodes must be 20 billion or higher. ⒠ File creation performance in a shared directory must be 100,000 files/sec or higher. ⒡ Each device of the parallel file system must be swappable without stopping the entire system. ⒢ The performance monitoring system for the parallel file system must be installed. C Login servers must satisfy the following conditions. ⒜ Login servers must consist of 3 nodes or more. ⒝ Login servers must be connected to the interconnection network of all compute nodes. ⒞ Each login server must have a network interface for external network at the rate of 40 Gbps or higher, and must be connected to an external network router. ⒟ The external network router must have 6 ports or more of 10 Gbit Ethernet. ⒠ Login servers must mount the parallel file system of Pegasus supercomputer via the InfiniBand network. D The software available on all the compute nodes must satisfy the followingconditions. ⒜ The system must be operated by Linux operating system. ⒝ For CPU, Fortran, C and C++ including automatic parallelization feature and OpenMP must be provided. ⒞ For GPU, CUDA (including CUDA translator) and OpenACC or OpenMP must be provided. C, C++ and Fortran must be provided. ⒟ MPI 4.0 or later must be provided for parallel processing over all compute nodes. ⒠ Highly parallelized numerical and artificial intelligence library must be provided. These libraries must be available in Python. ⒡ Batch-job scheduling system to manage all the compute nodes must be provided. E The power consumption of parallel computing servers, the parallel file system, login servers and all the cooling system must be equal to or less than 300 kVA in total. To support the continuous operation of parallel processing servers and the parallel file system which are fully powered, power supply unit capacity, cooling system, floor plan, and all the environments must be carefully considered. ⑸ Time limit for the submission of the requested material : 17 : 00 20 May, 2024 ⑹ Contact point for the notice : Takashi Matsuzaka Associate Head Center forComputational Sciences, University of Tsukuba 1-1-1 Tennodai Tsukuba-shi 305-8577 Japan TEL 029-853-6486 |
