Aggregation of t-digest types broken by default
See original GitHub issueIn practice, summary data in prometheus can be hard to aggregate, because they lose a bit of data. I’ve been using a custom IPC setup to push summary data generated on workers to a master’s summary metric to avoid integration. However, this is cumbersome and I’d like to use the default metrics aggregator provided by prom-client
. Can you speak to how this works in practice? It’s a bit difficult for me to figure out.
Issue Analytics
- State:
- Created 5 years ago
- Comments:12 (1 by maintainers)
Top Results From Across the Web
Need a more flexable histogram / tdigest · Issue #6440
It appears there are two modes for the aggregator. In the "local" mode a set of quantiles are requested and in the "central"...
Read more >tdigest() (aggregation function) - Azure Data Explorer
Expression used as the weight of the values for the aggregation calculation. Returns. The Intermediate results of weighted percentiles of *Expr* ...
Read more >Cisco ASR 9000 Series Aggregation Services Router ...
The default limits can be overridden through configuration of the maximum-prefix limit command for the peer for the appropriate address family. The following ......
Read more >pt-query-digest — Percona Toolkit Documentation
When parsing general logs ( --type genlog ), the default --order-by becomes Query_time:cnt . General logs do not report query times so only...
Read more >RFC 4271: A Border Gateway Protocol 4 (BGP-4)
BGP-4 also introduces mechanisms that allow aggregation of routes, ... Parameter Type is a one octet field that unambiguously identifies individual ...
Read more >
Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free
Top Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
Ah, I think you are correct. Here’s a test script:
And the output:
I think this is probably not the behavior people would expect with default aggregation. I believe that summaries and histograms used the t-digest primitive, which appears to be built to be combined. If so, it seems like we could send the t-digest data over when summaries need to be aggregated.
Alternatively, might it be better for the library to do something under-the-covers where the master worker holds the “real” in-memory instrument, and the worker nodes, when observing measurements, merely pass those measurements over IPC to the master?
I renamed the title to be a bit more on the mark and am bumping the thread! Any news?