feat(experiments): E4-pressured points at third_party/traces SWE-Bench trace
Switches the default --trace from outputs/inferact_50sess.jsonl (median 63K, p99 143K, 1285 reqs) to third_party/traces/qwen35-swebench-50sess.jsonl (median 27K, p99 92K, 4449 reqs across 52 sessions). Smaller per-request inputs let us check whether the queue-induced TTFT collapse the user flagged is workload-specific. Total trace is 3.5x larger so the run will cover more turns per session.
This commit is contained in:
@@ -26,7 +26,7 @@ if [ -z "${CUDA_HOME:-}" ]; then
|
|||||||
fi
|
fi
|
||||||
|
|
||||||
MODEL=${MODEL:-/mnt/models/Qwen/Qwen3-30B-A3B-Instruct-2507}
|
MODEL=${MODEL:-/mnt/models/Qwen/Qwen3-30B-A3B-Instruct-2507}
|
||||||
TRACE=${TRACE:-outputs/inferact_50sess.jsonl}
|
TRACE=${TRACE:-third_party/traces/qwen35-swebench-50sess.jsonl}
|
||||||
OUTPUT=${OUTPUT:-outputs/e4p_kvc_v2_d_to_p_sync_pressured_50sess}
|
OUTPUT=${OUTPUT:-outputs/e4p_kvc_v2_d_to_p_sync_pressured_50sess}
|
||||||
IB_DEVICE=${IB_DEVICE:-mlx5_60}
|
IB_DEVICE=${IB_DEVICE:-mlx5_60}
|
||||||
LOAD_FLOOR_BONUS=${LOAD_FLOOR_BONUS:-200}
|
LOAD_FLOOR_BONUS=${LOAD_FLOOR_BONUS:-200}
|
||||||
|
|||||||
Reference in New Issue
Block a user