version = 1 [paths] state_dir = "runs/frontier-tp2-prefill-serving-v0/fleet-state" artifacts_dir = "runs/frontier-tp2-prefill-serving-v0/fleet-artifacts" [ssh] connect_timeout_sec = 10 [scheduler] gpu_free_memory_mb = 1024 gpu_free_utilization_pct = 10 prefer_pack = true # Code is synchronized by Git. The fleet's mandatory rsync pass is intentionally # empty so it only schedules and harvests the committed remote checkout. [sync] mode = "rsync" local_path = "runs/frontier-tp2-prefill-serving-v0" exclude = ["*"] [[hosts]] name = "dash1" ssh_alias = "dash1" enabled = true sync_remote_path = "/home/admin/cpfs/wjh/aituner/aituner" fleet_root = "/home/admin/cpfs/wjh/aituner/gpu-fleet-tp2-prefill-serving-v0/dash1" [[hosts]] name = "dash2" ssh_alias = "dash2" enabled = true sync_remote_path = "/home/admin/cpfs/wjh/aituner/aituner" fleet_root = "/home/admin/cpfs/wjh/aituner/gpu-fleet-tp2-prefill-serving-v0/dash2" [[hosts]] name = "dash3" ssh_alias = "dash3" enabled = true sync_remote_path = "/home/admin/cpfs/wjh/aituner/aituner" fleet_root = "/home/admin/cpfs/wjh/aituner/gpu-fleet-tp2-prefill-serving-v0/dash3" [[hosts]] name = "dash4" ssh_alias = "dash4" enabled = true sync_remote_path = "/home/admin/cpfs/wjh/aituner/aituner" fleet_root = "/home/admin/cpfs/wjh/aituner/gpu-fleet-tp2-prefill-serving-v0/dash4"