microsoft/mscclpp
1
0
Fork 0
mirror of https://github.com/microsoft/mscclpp.git synced 2026-05-05 22:22:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Report undirectional bandwidth

Browse Source
This commit is contained in:
Qinghua Zhou
2026-02-24 06:02:33 +00:00
parent f803eff8b9
commit 6292b6ab33
1 changed files with 8 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
python/test/test_alltoallv_mscclpp.py
View File

@@ -157,8 +157,8 @@ def main():
# Test 3: Performance benchmark across message sizes (1KB to 128MB)
if rank == 0:
print("\n[Test 3] Performance benchmark (1KB to 128MB per rank)")
print(f" {'Msg Size':>10s} {'Iters':>5s} {'Total (ms)':>10s} {'Lat (us)':>10s} {'BW (GB/s)':>10s}")
print(f" {'-'*10} {'-'*5} {'-'*10} {'-'*10} {'-'*10}")
print(f" {'Msg Size':>10s} {'Iters':>5s} {'Total (ms)':>10s} {'Lat (us)':>10s} {'algBW(GB/s)':>12s}")
print(f" {'-'*10} {'-'*5} {'-'*10} {'-'*10} {'-'*12}")
# Message sizes: 1KB, 4KB, 16KB, 64KB, 256KB, 1MB, 4MB, 16MB, 64MB, 128MB
msg_sizes = [1 << s for s in range(10, 28) if s % 2 == 0] # powers of 4 from 1KB to 64MB
@@ -187,7 +187,11 @@ def main():
torch.cuda.synchronize()
elapsed = time.perf_counter() - start
total_bytes = 2 * input_size * n_iters # read + write
# Algorithm bandwidth: total data received per rank / time
# Each rank receives msg_size bytes from each of the other (world_size-1) peers
# plus msg_size from itself (local copy). Total recv = input_size = msg_size * world_size.
# Report unidirectional algBw (same convention as nccl-test).
total_bytes = input_size * n_iters
bandwidth_gbps = total_bytes / elapsed / 1e9
latency_us = elapsed / n_iters * 1e6
@@ -198,7 +202,7 @@ def main():
size_str = f"{msg_size // 1024}KB"
else:
size_str = f"{msg_size}B"
print(f" {size_str:>10s} {n_iters:>5d} {elapsed*1000:>10.2f} {latency_us:>10.1f} {bandwidth_gbps:>10.2f}")
print(f" {size_str:>10s} {n_iters:>5d} {elapsed*1000:>10.2f} {latency_us:>10.1f} {bandwidth_gbps:>12.2f}")
# Cleanup
dist.barrier()