DDBJ Database updates and computational infrastructure enhancement.

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• The Bioinformation and DDBJ Center (DDBJ) at the National Institute of Genetics (NIG) is a primary nucleotide sequence database and a member of the International Nucleotide Sequence Database Collaboration (INSDC).
• NIG operates a supercomputer essential for DDBJ database construction and serves as a large-scale computational resource for Japanese researchers.

Purpose of the Study:

• To address the exponential growth of deoxyribonucleic acid (DNA) sequence data.
• To enhance computational infrastructure for big data analysis in genomics and bioinformatics.

Main Methods:

• Replacement of the NIG supercomputer system in early 2019 with a new system designed for big data analysis.
• Installation of 30 PB DNA data archiving storage, 13.8 PB of parallel distributed file systems, and 1.1 PFLOPS computation nodes with graphics processing units (GPUs).
• Implementation of an automatic file transfer system to develop a multi-cloud infrastructure for bioinformatics.

Main Results:

• The new supercomputer system significantly increases data storage and computational power for genome data analysis.
• The multi-cloud infrastructure enables users to leverage optimal computing environments (supercomputer or public clouds) for different workloads.
• The automatic file transfer system facilitates data portability, preventing data lock-in and balancing cost/performance.

Conclusions:

• The upgraded NIG supercomputer and multi-cloud infrastructure provide a robust and flexible platform for advancing bioinformatics and genomics research.
• This initiative supports Japanese biologists and medical researchers by offering enhanced computational resources and efficient data management solutions.
• The development of a multi-cloud strategy positions NIG at the forefront of bioinformatics infrastructure, promoting innovation and collaboration.