ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 Performance Evaluation of Fabric-Attached Memory Pool for AI Applications
Cited 0 time in scopus Download 2 time Share share facebook twitter linkedin kakaostory
저자
김영우, 오명훈
발행일
202102
출처
International Conference on Electronics, Information and Communication (ICEIC) 2021, pp.586-589
DOI
https://dx.doi.org/10.1109/ICEIC51217.2021.9369733
협약과제
20HS2700, 메모리 중심 차세대 컴퓨팅 시스템 구조 연구, 오명훈
초록
Recently, traditional compute-oriented system architecture is changing and diverting. The increasing demands for big data and artificial intelligence accelerate the needs for new and alternative architecture in computing system. One of the emerging area for alternative computing architecture is memory-centric or disaggregated computing. The memory-centric or disaggregated computing can solve the requirement for huge memory in system wide. In this paper, we present preliminary performance evaluation results of the fabric-attached memory system with industry standard based memory pool prototype hardware. The memory pool prototype hardware is configured as block device for benchmarking. For evaluation, well-known benchmarks - sysbench and resnet - are used. The preliminary benchmark results show that the overall access performance of the fabric-attached prototype hardware is comparable to SSD, and deep learning performance is close to NVMe.
키워드
Fabric-attached memory, Gen-Z protocol, Memory centic, Resnet
KSP 제안 키워드
AI Applications, Benchmark results, Big Data, Block device, Disaggregated computing, Industry standard, Learning performance, Memory System, Memory-centric, Performance evaluation, System architecture