Professor Nagasaki and Professor Kawaguchi of Center for Genomic Medicine, Kyoto University Graduate School of Medicine

High-performance TS-2483XU-RP accelerates and streamlines genome data analysis.

  • Background

    Professor Nagasaki and Professor Kawaguchi of the Center for Genomic Medicine, Kyoto University Graduate School of Medicine study genomes, metabolites, and health information searching for disease-related factors in humans as their main goal. They conduct large scale data analysis of tens of thousands of people.

  • Challenges

    The genome consists of 3 billion base pairs (3GB) per person, and when analyzing the genome, an extreme amount of data must be processed. Therefore, high-performance storage was required. The professors previously used RAID storage directly connected to each computing server as on-premises near-line storage to accommodate all data. With this method, I/O speed was a bottleneck, and the data was literally scattered across several storage units, making it difficult for the researchers to utilize it.

Solution

The professors installed three QNAP TS-2483XU-RP for high I/O speed with post-warranty parts availability, cost efficiency and easier portability (when they require physically relocating the NAS). The TS-2483XU-RP is a high-performance 24-bay NAS powered by Intel ® Xeon® E processors with up to 128GB DDR4 ECC memory and including two 10GbE SFP+ ports. The professors also installed other high-speed distributed storage and integrated them with near-line storage consisting of QNAP NAS to separately store data based on its purpose. The analysis data is stored on the fastest storage, and base data is stored as cold data, which is essential but infrequently accessed.

Results

The professors got a complete suite of QNAP storage, enabling them to greatly reduce both learning and management cost with a centralized management interface and GUI-based configuration. QNAP NAS is easy to install and has become indispensable as a medium for exchanging data with external institutions. It enabled them to improve user convenience through centralized data access and build a cost-efficient storage system that meets I/O requirements. The professors are building a hybrid cloud environment that combines three infrastructures: on-premises servers, public clouds, and interdisciplinary High Performance Computing Infrastructure (as known as supercomputers) on the SINET* academic information network. Not only as the near-line storage originally envisioned, but the professors also use QNAP NAS as a direct I/O environment from computer clusters. In the future, the professors aim to create a more seamless environment in which the data of each platform is centralized, and users can utilize desired data without being aware of the differences in platforms and applications.

*SINET is the academic IT network infrastructure for Japanese universities and research institutes.

Teknik Özellik sSçin

      Daha fazla göster Daha az

      Diğer ülkelerde/bölgelerde bu site:

      open menu
      back to top