[AMD] dsv4-fp4-mi355x-atom: enable DPA TBO at high concurrency, update image to atom0.1.4 #1717
+56
−22
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -22,31 +22,51 @@ echo "TP: $TP, CONC: $CONC, ISL: $ISL, OSL: $OSL, EP_SIZE: $EP_SIZE, DP_ATTENTIO | |
| SERVER_LOG=/workspace/server.log | ||
|
|
||
| PARALLEL_ARGS=(-tp "$TP") #TP | ||
| CUDAGRAPH_SIZES='[1,2,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140,144,148,152,156,160,164,168,172,176,180,184,188,192,196,200,204,208,212,216,220,224,228,232,236,240,244,248,252,256]' | ||
| if [ "$DP_ATTENTION" = "true" ]; then | ||
| if [ "$EP_SIZE" -gt 1 ]; then #DP+EP | ||
| PARALLEL_ARGS=(-tp "$TP" --enable-expert-parallel --enable-dp-attention ) | ||
| else #DPA+TP | ||
| #DPA+TP+TBO | ||
| if [ "$ISL" -eq 1024 ] && [ "$OSL" -eq 1024 ] && [ "$CONC" -ge 1024 ]; then | ||
| PARALLEL_ARGS=(-tp "$TP" --enable-dp-attention --enable-tbo) | ||
| CUDAGRAPH_SIZES='[1,2,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140,144,148,152,156,160,164,168,172,176,180,184,188,192,196,200,204,208,212,216,220,224,228,232,236,240,244,248,252,256,512,1024]' | ||
| elif [ "$ISL" -eq 8192 ] && [ "$OSL" -eq 1024 ] && [ "$CONC" -ge 256 ]; then | ||
| PARALLEL_ARGS=(-tp "$TP" --enable-dp-attention --enable-tbo) | ||
| else | ||
|
There was a problem hiding this comment. CUDA graph sizes omit high concurrencyMedium Severity For ISL=8192 and OSL=1024,   Reviewed by Cursor Bugbot for commit 7ffa976. Configure here. |
||
| PARALLEL_ARGS=(-tp "$TP" --enable-dp-attention ) | ||
| fi | ||
| fi | ||
| fi | ||
|
|
||
| BENCHMARK_MAX_MODEL_LEN="$MAX_MODEL_LEN" | ||
|
|
||
| if [ "${EVAL_ONLY}" = "true" ]; then | ||
| EVAL_MAX_MODEL_LEN=$(compute_eval_context_length "$MODEL" "$BENCHMARK_MAX_MODEL_LEN") | ||
| export EVAL_MAX_MODEL_LEN | ||
| SERVE_MAX_MODEL_LEN="$EVAL_MAX_MODEL_LEN" | ||
| else | ||
| SERVE_MAX_MODEL_LEN="$BENCHMARK_MAX_MODEL_LEN" | ||
| fi | ||
| # Start GPU monitoring (power, temperature, clocks every second) | ||
| start_gpu_monitor | ||
|
|
||
| set -x | ||
| export ATOM_DISABLE_MMAP=true | ||
| export AITER_BF16_FP8_MOE_BOUND=0 | ||
| export ATOM_MOE_GU_ITLV=1 | ||
| python3 -m atom.entrypoints.openai_server \ | ||
| --model $MODEL \ | ||
| --server-port $PORT \ | ||
| "${PARALLEL_ARGS[@]}" \ | ||
| --kv_cache_dtype fp8 \ | ||
| --trust-remote-code \ | ||
| --gpu-memory-utilization 0.85 \ | ||
| --no-enable_prefix_caching \ | ||
| --cudagraph-capture-sizes "${CUDAGRAPH_SIZES}" \ | ||
| --max-num-seqs ${CONC} \ | ||
| > $SERVER_LOG 2>&1 & | ||
| #--max-model-len "$SERVE_MAX_MODEL_LEN" \ | ||
|
There was a problem hiding this comment. Eval context not wiredMedium Severity This commit adds Reviewed by Cursor Bugbot for commit c3b3289. Configure here. |
||
|
|
||
| SERVER_PID=$! | ||
|
|
||
|
|
||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Search space not enabled
High Severity
The
dsv4-fp4-mi355x-atomfixed-seq-len matrix no longer matches the PR: the ISL=1024 scenario is fully commented out, and ISL=8192 only runs a single DPA point at conc=1024 instead of the documented TP8 (conc 4–64) and DPA (conc 128–1024) sweeps.Reviewed by Cursor Bugbot for commit c3b3289. Configure here.