Computing Platform, focus on large-scale bioinformatics processing, continually improves high performance computing and cloud computing to provide innovative solution for huge amounts of data.

BGI has several bioinformatics data centers, located in Shenzhen, Hong Kong, Beijing and Wuhan respectively, with the total peak performance up to 373.3 T flopsthe total memory capability up to 90.22 TB and total storage capability up to 44.84 PB. Among these, data center in shenzhen and in hongkong ranks the first and second respectively in the bioinformatics field in China.They provide stable and efficient resource for storing, processing and analyzing the massive bioinformatics data.


Site

CPU Cores

Memory(TB)

Storage(PB)

T flops 

Shenzhen

18,200

52

22.74

199

Hong Kong

8,336

12

14.47

93

Beijing

300

0.5

0.5

1.5

Wuhan

4,944

19.9

5.33

54

Tianjin

632

2.42

0.65

14.2

CHOPS

300

0.8

0.5

2

Denmark

920

2.6

0.65

9.6

Total

33,632

90.22

44.84

373.3

BGI Computing Platform

With the advancements of gene research technologies, big data processing makes a high requirement for computing and storage capabilities increasing by 10-fold every 12-18 months, far above the Moore law’s reference value. Under the background of big data bang, BGI continues increasing the investment of hardware resource. In 2014, the first BGI’s supercomputing cluster with the computing capability over 1PFLOPS will come into service and become the Performance benchmarking in the bioinformatics research field of china, even in the world.

BGI, with the ability of world’s leading bioinformatics computing and analyzing, has worked with national supercomputing center to build the TH-BGI bioinformatics union lab located in national supercomputing center base-binhai new area in Tianjin, with an aim to create a world’s leading gene data computing and development base. The cooperation fully leverages the TH-1’s computing capability, starting with developing high performance computing application technology, including optimizing the existing bioinformatics software algorithm and developing high-quality bioinformatics analysis tools to resolve the issues massive amounts of data processing faces.

   

         Figure 1. TH-1A supercomputer                                    Figure 2. TH-1 system