Kanehisa Laboratories

Fukuoka - Kyoto - Tokyo

History

The Kanehisa Laboratory in the Institute for Chemical Research (ICR), Kyoto University was established in October 1985 when Minoru Kanehisa returned from the USA. In the late 1980s he became involved in a pilot study of the Human Genome Project in Japan, and started his efforts to introduce a supercomputer system at ICR.

In 1991 the Japanese Human Genome Program was officially started and the Human Genome Center (HGC) was established in the Institute of Medical Science, University of Tokyo. The Laboratory of Genome Database was the first laboratory of HGC and Kanehisa became chief of the Laboratory.

In the same year ICR introduced the first supercomputer system, which became operational in January 1992. The ICR Supercomputer System shown below has played essential roles in the development of the GenomeNet service since 1992 and the KEGG database project since 1995.

ICR Supercomputer System

1992.1-1996.12 Cray Y-MP2E/264, 2 CPUs, 512 MB
Sun SPARCserver 490, 64 MB
1997.1-2001.12 Cray T94/4128, 4 CPUs, 1 GB
Cray Origin 2000, 128 CPUs, 16 GB
Sun Enterprise 3000, 6 CPUs, 1 GB
2002.1-2007.12 SGI Origin 3800, 512 CPUs, 512 GB
SGI Origin 3800, 256 CPUs, 256 GB
Sun Fire 15K, 72 CPUs, 144 GB (3 sets)
2008.1-2011.12 SGI Altix 4700, 512 cores, 2.2 TB
SGI Altix 4700, 256 cores, 1.1 TB
Sun Fire E6900 (4 sets), total of 144 cores, 576 GB
Sun Fire X4600 (3 sets), total of 40 cores, 80 GB
2012.1-2015.12 SGI UV 1000, 1024 cores, 16 TB (2 sets)
SGI UV 1000, 512 cores, 8 TB (2 sets)
Oracle Sun Fire X4800, 64 cores, 512 GB (12 sets)
2016.1-2019.12 SGI UV2000, 512 cores, 16 TB (2 nodes)
SGI C2112, 20 cores, 256 GB (150 nodes)
Dell PowerEdge R930, 64 cores, 512 GB (10 nodes)
2020.1-2023.12 HPE Superdome Flex, 576 cores, 24 TB (2 nodes)
HPE Apollo 2000, 40 cores, 384 GB (142 nodes)
HPE ProLiant DL560, 64 cores, 768 GB or 1.5 TB (10 nodes)
2024.1- HPE Superdome Flex, 576 cores, 24 TB (2 nodes)
HPE Apollo 2000, 56 cores, 256 GB or 512 GB or 1 TB (111 nodes)
HPE ProLiant DL560, 64 cores, 768 GB or 1.5 TB (10 nodes)