Skip to content

storage_tokens_processed undercounts by tensor_parallel factor #402

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
FileSystemGuy wants to merge 7 commits into
base: main
Choose a base branch
from
+2 −1
Open

storage_tokens_processed undercounts by tensor_parallel factor #402

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
e6b4eee
storage_tokens_processed undercounts by tensor_parallel factor
FileSystemGuy Jun 2, 2026
b5a6028
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 11, 2026
f68334f
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 11, 2026
39ebad2
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 12, 2026
d15bb96
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 12, 2026
033430d
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 13, 2026
8d150fa
Merge branch 'main' into FileSystemGuy-kvcache-storage_tokens_processed
FileSystemGuy Jun 13, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion kv_cache_benchmark/kv_cache/cache.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -826,7 +826,8 @@ def access_cache(self, key: str, phase: InferencePhase = InferencePhase.DECODE,
self.stats['storage_read_host_latencies'].append(timing.host)

if self.model_config.kv_cache_size_per_token > 0:
num_tokens = entry_size / self.model_config.kv_cache_size_per_token
sharded_bytes_per_token = kv_cache_size_per_token / max(1, self.tensor_parallel)
num_tokens = entry_size / sharded_bytes_per_token
self.stats['storage_tokens_processed'] += num_tokens

return location, timing.total
Expand Down
Loading