nodes_with_vram (view)

20 rows

✎ View and edit SQL

This data as json, CSV (advanced)

Suggested facets: PARTITION, TIMELIMIT, NODES, STATE, VRAM

PARTITION AVAIL TIMELIMIT NODES STATE NODELIST VRAM
p4000 up 1-00:00:00 15 alloc c[0-1,3,8-19] 8
p4000 up 1-00:00:00 1 idle c2 8
rtx2060 up 1-00:00:00 6 alloc c[26,53-57] 6
rtx2060 up 1-00:00:00 6 idle c[29,51-52,60-62] 6
rtx2070 up 1-00:00:00 5 alloc c[69-73] 8
rtx2070 up 1-00:00:00 10 idle c[50,63-67,74-76,79] 8
rtx2080 up 1-00:00:00 2 idle c[24,28] 8
rtx2060super up 1-00:00:00 1 drain c38 8
rtx2060super up 1-00:00:00 15 alloc c[21,25,32,34,39-42,44-47,58-59,68] 8
rtx2060super up 1-00:00:00 5 idle c[37,43,48-49,78] 8
rtx2080super up 1-00:00:00 1 alloc c22 8
rtx2080super up 1-00:00:00 1 idle c23 8
rtx3060ti up 1-00:00:00 2 alloc c[27,77] 8
a4000 up 1-00:00:00 3 alloc c[4-5,35] 16
a4000 up 1-00:00:00 3 idle c[6-7,36] 16
a6000 up 8-00:00:00 2 alloc c[30-31] 48
a100 up 1-00:00:00 1 alloc c33 80
rtx4060ti8g up 1-00:00:00 5 alloc c[20-22,26-27] 8
rtx4060ti8g up 1-00:00:00 2 idle c[28-29] 8
rtx4060ti16g up 1-00:00:00 1 alloc c25 16

Advanced export

JSON shape: default, array, newline-delimited

CSV options: 

CREATE VIEW nodes_with_vram AS SELECT nodes.*, [nvidia-gpus].VRAM
FROM nodes
INNER JOIN [nvidia-gpus] ON nodes.PARTITION = [nvidia-gpus].name;