본문 바로가기 대메뉴 바로가기
open
Close

Super Computing center

인프라

AI X Cluster

AI-X Cluster Architecture

  • Management Network(10G)
  • Data Network(100G RoCE)
  • Internal Network(100G 1B)
  • Internal Network(200G 1B)
Front Cluster-(controller + Master) AI-X Data Pond 140 TB: Each user ~ 3TB AI-X Data Lake 170 TB: Each user ~ 5TB Computer + DataPond Nodes(2u4N) Computer + Datapond Node 1(6x 6.4TB), Computer + Datapond Node 2(6x 6.4TB), Computer + Datapond Node 3(6x 6.4TB), Computer + Datapond Node 4(6x 6.4TB). Mellanox SN2100 (100G RoCE) / Backend Cluster-Core Cloud (Slurm + K85 Master) Local Storage 3 TB: Each user ~ 100GB. AI-X Data Lake 170 TB: Each user ~ 5TB. Storage Nodes (AI-X Data Lake). FlashBlade 17TB X 10. External Fabric Module (XFM). GPU Node 1 (DGXA100-1) AI Graduate School-1-node Multi-GPUs (8x A100), GPU Node 2 (DGXA100-2) AI Graduate School-1-node Multi-GPUs (8x A100), GPU Node 3 (DGX1v-1) AI Graduate School-1-node Multi-GPUs (8x Tesla V100), GPU Node 4 (DGX1v-2) AI Graduate School-1-node Multi-GPUs (8x Tesla V100). Mellanox SN2100 (100G RoCE), Mellanox QM8700 (200G IB)
큰 이미지로 보기 +

Computing

DGX A100

  • DGX A100
  • 8x NVIDIA A100 GPUs
  • 320GB GPU Memory
  • 1TB System Memory
  • Dual AMD Rome 2.25GHz(64-core) CPU
  • 9x 200Gb/s NIC
DGX A100

DGX-1V

  • 8x NVIDIA V100 GPUs
  • 256GB GPU Memory
  • 512GB System Memory
  • Dual Intel Xeon E5-2698 v4 2.2GHz(20-core) CPU
  • 4x 100Gb/s NIC
DGX-1V

Storage

Network File System(NFS v3)

  • NFS server and client

Ceph

  • Software-defined storage ( Object storage, Block storage, File storage)

Networking

AI-X Computing Cluster

  • Data network : Mellanox SN2100 (100G RoCE)
Mellanox SN2100
  • Internal network : Mellanox QM8700 (200G Infiniband)
Mellanox QM8700

Commonly used in HPC

  • 높은 처리량과 짧은 대기시간
  • 슈퍼컴퓨터와 스토리지 시스템 연결
  • RoCE : 통합 이더넷을 통한 RDMA
  • Infiniband(IB) : SAN(System Area Network)을 위한 짧은 대기 시간 및 고대역폭

Link speed

  • Enhanced Data Rate(EDR) - 25Gb/s per lane (100Gb/s for 4x)
  • High Data Rate(HDR) - 50Gb/s per lane (200Gb/s for 4x)
AI-X Cluster 장비1
AI-X Cluster 장비2
QUICK
MENU
꿈꾸는 아이꿈꾸는 아이 Ai 대학원Ai 대학원 GIST 대표GIST 대표