Skip to content
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

Jump to bottom

Decode Batch Percentage Metrics/Improved Scraping #82

Merged
merged 77 commits into from
May 20, 2024
Merged

Decode Batch Percentage Metrics/Improved Scraping #82

merged 77 commits into from
May 20, 2024

Conversation

Bslabe123
Copy link
Contributor

@Bslabe123 Bslabe123 commented May 13, 2024 •
edited

Changes:

  • Added metric for available slots in decode batch percentage
  • Changed scraping technique for prefill backlog queue metric
  • Prometheus client endpoint enablement/port not set via ENV anymore

Results from various prometheus client queries:

# HELP jetstream_prefill_backlog_size Size of prefill queue
# TYPE jetstream_prefill_backlog_size gauge
jetstream_prefill_backlog_size{id="maxengine-server-58f8786f4c-4587n"} 57.0
# HELP jetstream_slots_available_percentage The percentage of available slots in decode batch
# TYPE jetstream_slots_available_percentage gauge
jetstream_slots_available_percentage{id="maxengine-server-58f8786f4c-4587n",idx="0"} 0.78125

# HELP jetstream_prefill_backlog_size Size of prefill queue
# TYPE jetstream_prefill_backlog_size gauge
jetstream_prefill_backlog_size{id="maxengine-server-58f8786f4c-4587n"} 28.0
# HELP jetstream_slots_available_percentage The percentage of available slots in decode batch
# TYPE jetstream_slots_available_percentage gauge
jetstream_slots_available_percentage{id="maxengine-server-58f8786f4c-4587n",idx="0"} 0.375

# HELP jetstream_prefill_backlog_size Size of prefill queue
# TYPE jetstream_prefill_backlog_size gauge
jetstream_prefill_backlog_size{id="maxengine-server-58f8786f4c-4587n"} 0.0
# HELP jetstream_slots_available_percentage The percentage of available slots in decode batch
# TYPE jetstream_slots_available_percentage gauge
jetstream_slots_available_percentage{id="maxengine-server-58f8786f4c-4587n",idx="0"} 0.96875

@Bslabe123 Bslabe123 marked this pull request as ready for review May 14, 2024 22:51
FanhaiLu1
FanhaiLu1 reviewed May 17, 2024
View reviewed changes
Copy link
Collaborator

@FanhaiLu1 FanhaiLu1 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

new version of code looks good to me

JoeZijunZhou
JoeZijunZhou reviewed May 17, 2024
View reviewed changes
requirements.txt Outdated Show resolved Hide resolved
@JoeZijunZhou JoeZijunZhou merged commit 44686b0 into google:main May 20, 2024
3 checks passed
@Bslabe123 Bslabe123 deleted the prometheus-decode-batch-percent branch May 20, 2024 18:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@JoeZijunZhou JoeZijunZhou JoeZijunZhou approved these changes

@FanhaiLu1 FanhaiLu1 FanhaiLu1 approved these changes

@vipannalla vipannalla Awaiting requested review from vipannalla vipannalla is a code owner

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@Bslabe123 @JoeZijunZhou @FanhaiLu1