Add n_probe to the output of :qa:vector:checkVec and allow multiple n_probe in the same run
#130316
Conversation
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
iverase
added
>non-issue
:Search Relevance/Search
Collaborator
|
Pinging @elastic/es-search-relevance (Team:Search Relevance) |
elasticsearchmachine
added
the
Team:Search Relevance
benwtrent
approved these changes
Jun 30, 2025
iverase
added
the
auto-merge-without-approval
mridula-s109
pushed a commit
to mridula-s109/elasticsearch
that referenced
this pull request
Jul 3, 2025
…ic#130316) This change adds the n_probe value to the output which will be 0 in the case of non-ivf runs. In addition it separates index and search data, so a normal output looks like: ``` index_type num_docs index_time(ms) force_merge_time(ms) num_segments ---------- -------- -------------- -------------------- ------------ ivf 1000000 50382 132819 0 index_type n_probe latency(ms) net_cpu_time(ms) avg_cpu_count QPS recall visited ---------- ------- ----------- ---------------- ------------- ------ ------ -------- ivf 100 3.69 0.00 0.00 271.00 0.97 58917.00 ``` In addition, this change allows to define an array of n_probe in the configuration file so we can test different values in the same run, so for example defining an n_probe like: ``` "n_probe" : [10, 20, 30, 40, 50, 60, 70, 80, 90, 100], ``` will produce the following output: ``` index_type num_docs index_time(ms) force_merge_time(ms) num_segments ---------- -------- -------------- -------------------- ------------ ivf 1000000 50382 132819 0 index_type n_probe latency(ms) net_cpu_time(ms) avg_cpu_count QPS recall visited ---------- ------- ----------- ---------------- ------------- ------ ------ -------- ivf 10 1.18 0.00 0.00 847.46 0.82 7244.59 ivf 20 1.36 0.00 0.00 735.29 0.89 13288.69 ivf 30 1.66 0.00 0.00 602.41 0.92 19266.67 ivf 40 1.93 0.00 0.00 518.13 0.94 24995.41 ivf 50 2.21 0.00 0.00 452.49 0.94 30739.60 ivf 60 2.51 0.00 0.00 398.41 0.95 36428.00 ivf 70 2.76 0.00 0.00 362.32 0.96 41952.59 ivf 80 2.99 0.00 0.00 334.45 0.96 47599.64 ivf 90 3.31 0.00 0.00 302.11 0.96 53254.45 ivf 100 3.69 0.00 0.00 271.00 0.97 58917.00 ``` This makes easier to plot the n_probe curve while doing changes.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
This change adds the n_probe value to the output which will be 0 in the case of non-ivf runs. In addition it separates index and search data, so a normal output looks like:
In addition, this change allows to define an array of n_probe in the configuration file so we can test different values in the same run, so for example defining an n_probe like:
will produce the following output:
This makes easier to plot the n_probe curve while doing changes.