A domestic research team has developed a computing system technology that can perform genome analysis even faster, improving analysis performance by 28% compared to existing processor-centric systems.

The Electronics and Telecommunications Research Institute (ETRI) announced on the 23rd that it has developed a memory-centric computing system specialized for genome analysis. This technology has achieved a 28% performance improvement over existing technologies. If the service time previously took about 10 months, this can be shortened to about 7 months.

The technology developed by ETRI is a memory-centric computing hardware and software technology specialized in next-generation sequencing (NGS) that analyzes genomes.

Until now, genome analysis has mainly used processor-centric computing technology that uses limited memory but performs many calculations. When processing large amounts of data such as genome analysis, structural bottlenecks often occur, requiring a lot of time and effort for data processing.

On the other hand, ETRI's memory-centric computing technology overcomes bottlenecks by utilizing large-scale memory.

First, the research team developed a hardware device called MOCA that enabled the system to be equipped with large-scale memory.

The key is to eliminate the need to use hard disks or SSDs during the data processing process.

Additionally, the research team developed a software that can process the base sequence alignment step, which takes the longest time during the genome analysis process, more than twice as fast by utilizing large-scale memory, thereby increasing analysis efficiency.

ETRI also verified the technology's performance in collaboration with GC Green Cross Genome.

As a result, it was shown that applying the HW developed by the research team to the existing system could increase the overall analysis performance by 16%, and applying HW and SW simultaneously could improve the performance by up to 28%.

This technology can be applied to systems that can predict cancer incidence, fetal defects, and other factors, as well as identify mutations in infectious diseases and develop treatments.

Thanks to this, it is expected that analytical institutions and pharmaceutical companies will be able to reduce service development costs and diagnosis times, and hospitals will be able to establish patient-tailored collaborative treatment systems, which will greatly contribute to improving national health and reducing social burden.

Kim Kang-ho, head of the ETRI Data-Centric Computing Systems Lab, said, “This technology is expected to serve as a new catalyst for the domestic pharmaceutical analysis market and industry, significantly activating various bio-application markets and creating employment.”