List of queues

Queues of the accelerator server are as follows.

Execlusive-Queue

Queue name Limit of number of exclusive nodes (default) Memory Limit [GiB] Walltime limit[hours](default) Limit of concurrent execution Upper Limit of using GPU Limit of number of parallel execution Notes
IA_001g 1(1) 690 24(1) 6 10 36 for interactive
DA_002g 2(1) 1,380 2(2) 6 20 72 for debugging
A_004 4(1) 2,760 72(24) No limit 40 144
IC_001 1(1) 510 24(1) No limit - 36 for interactive
DC_002 2(1) 1,020 2(2) No limit - 72 for debugging
C_002 2(1) 1,020 72(24) No limit - 72
C_004 4(1) 2,040 72(24) No limit - 72


Shared-Queue

Queue name Limit of number of CPU (default) Limit of number of GPU (default) Memory Limit [GiB] Walltime limit[hours] (default) Limit of concurrent execution Limit of number of parallel execution Notes
CA_001 18(1) 5(1) 345(69) 72(24) 10 18 Interactive mode available
CA_001g 18(1) 5(1) 345(69) 72(24) 10 18 Interactive mode available
CC_001 18(1) - 252(14) 72(24) No limit 18 Interactive mode available

important

To use the queue IA_001g, CA_001g and DA_002g, please submit a job from /work_da area.


We support researchers who want to develop application software for ultra-large-scale calculations on MASAMUNE-IMR and a new methodology for Computational Materials Science. We also offer following job queues for these calculations and you can use them by submitting your application. As for the application, please contact us using the contact form.

Queue name Limit of number of exclusive nodes (default) Memory Limit [GiB] Walltime limit[hours](default) Limit of concurrent execution Upper Limit of using GPU Limit of number of parallel execution Notes
LA_004 4(4) 2,760 168(96) 1 40 144 executed during operation
MA_008 8(8) 5,520 72(24) 1 80 288
SA_016 16(16) 11,040 72(72) 1 160 576 executed after maintenance operation only


  • user_manual/acceleratorserver/queues_list.txt
  • Last modified: 2023/09/19 08:12
  • by ccms