SYSTEM HISTORY

Date
History

2019-07-01

Official service start

2019-11-12

Install CUDA 9.0/9.1/10.0

2020-04-08

Expansion of 11 V100 GPU cards

2020-10-14

5 V100 8GPU/2GPU equipped servers, 9 V100 GPU cards installed

(New queue cas_v100nv_8, cas32c_v100_2)

2021-04-14

NVIDIA driver upgrade (v.440 → v.460)

2021-06-01

Install CUDA 10.1/10.2/11.2

2021-08-11

Apply Fairshare Scheduling

Full node OS upgrade (CentOS7.4 → 7.9)

2022-03-17

SLURM scheduler upgrade (v.18.08 → v.21.08)

Apply shared node policy

21 2GPU servers retired

(ivy_v100_2/cas_v100_2/cas32c_v100_2 queue retired and cas_v100_4 queue created)

2022-04-05

Install CUDA 11.0/11.1/11.3/11.4/11.5/11.6

2022-04-13

NVIDIA driver upgrade (v.460 → v.510)

2022-08-11

6/2/2 servers with A100 8 GPU/4 GPU/2 GPU (New queue amd_a100nv_8, amd_a100_4, amd_a100_2)

Added 2PB of HPE ClusterStor storage (DDN storage to be retired)

Last updated