Skip to content

Add a time series aggregation to tsdb track #348

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

Closed
martijnvg wants to merge 5 commits into from
Closed

Add a time series aggregation to tsdb track #348

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
62de908
Add a time series aggregation to tsdb track
martijnvg Nov 22, 2022
3315660
iter
martijnvg Nov 23, 2022
300a2c9
Increase search.max_buckets setting,
martijnvg Nov 24, 2022
549d1cd
adjust max buckets
martijnvg Nov 29, 2022
d49870e
300k is the right max buckets for time_series agg
martijnvg Nov 29, 2022
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
5 changes: 5 additions & 0 deletions tsdb/challenges/default.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,6 +123,11 @@
"operation": "date-histo-entire-range",
"warmup-iterations": 50,
"iterations": 100
},
{
"operation": "date-histo-with-time-series-1h",
"warmup-iterations": 50,
"iterations": 100
}
]
},
Expand Down
45 changes: 44 additions & 1 deletion tsdb/operations/default.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,7 +74,11 @@
{
"name": "date-histo-entire-range",
"operation-type": "search",
{%- if ingest_mode is defined and ingest_mode == "data_stream" %}
"index": "k8s",
{%- else %}
"index": "tsdb",
{%- endif %}
"body": {
"size": 0,
"aggs": {
Expand All @@ -86,4 +90,43 @@
}
}
}
}
},
{
"name": "date-histo-with-time-series-1h",
"operation-type": "search",
{%- if ingest_mode is defined and ingest_mode == "data_stream" %}
"index": "k8s",
{%- else %}
"index": "tsdb",
{%- endif %}
"body": {
"size": 0,
"aggs": {
"by_timestamp": {
"date_histogram": {
"field": "@timestamp",
"fixed_interval": "1h"
},
"aggs": {
"ts": {
"time_series": {
"keyed": false
},
"aggs": {
"available_memory": {
"min": {
"field": "kubernetes.node.memory.available.bytes"
}
}
}
},
"min_available_memory": {
Copy link
Contributor

@salvatore-campagna salvatore-campagna Nov 24, 2022

Choose a reason for hiding this comment

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

I am not sure why we are also including this pipeline aggregation. Do we expect to see changes in performance if we introduce some optimization for time_series aggregator?

Copy link
Member Author

@martijnvg martijnvg Nov 24, 2022

Choose a reason for hiding this comment

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

I added this because I think a pipeline aggregator is part of the usage pattern with time series aggregation.
All the time series buckets is an information overflow, getting min, max or derivative from all time series buckets will be useful informations to plot in a graph.

Copy link
Member Author

@martijnvg martijnvg Nov 24, 2022

Choose a reason for hiding this comment

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

Or do you think we should just focus on the time series agg itself in the benchmark? So removing the pipeline agg and maybe even the date histogram here? So it is just the time_series agg?

"min_bucket": {
"buckets_path": "ts>available_memory"
}
}
}
}
}
}
}