Skip to content

Memory Store: change default bufferSize from 1k to 100k #70

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
Closed
kimchy opened this issue Mar 18, 2010 · 1 comment
Closed

Memory Store: change default bufferSize from 1k to 100k #70

kimchy opened this issue Mar 18, 2010 · 1 comment
Labels
>enhancement v0.06.0

Comments

@kimchy
Copy link
Member

kimchy commented Mar 18, 2010

No description provided.

@kimchy
Copy link
Member Author

kimchy commented Mar 18, 2010

Memory Store: change default bufferSize from 1k to 100k, closed by b1e5284

@bluelu bluelu mentioned this issue Dec 11, 2014
Closed
dadoonet added a commit that referenced this issue Jun 5, 2015
@dadoonet
Remove deprecated settings
7ba6595 
#61 and #62 renamed settings by deprecating old ones and creating new ones.
In version 3.0.0, we can remove old settings.

That said, we should not remove it from master doc before 2.6.0 is released.

Closes #70.
@makeyang makeyang mentioned this issue Sep 7, 2015
Closed
maabernethy pushed a commit to maabernethy/elasticsearch that referenced this issue Mar 20, 2017
@kblincoe
Merge pull request elastic#70 from fernandozhu/master
4f2f025 
Claim issue 23570
henningandersen pushed a commit to henningandersen/elasticsearch that referenced this issue Jun 4, 2020
@danielmitterdorfer
Avoid more time-consuming calls (elastic#70)
8efa5cd 
With this commit we reduce usage of two time-consuming methods on the hot code
path: generating random numbers and determining the current timestamp. This also
changes behavior slightly:

* The `hostname` field will vary host names from 1 - 3 in a regular pattern
instead of randomly (however, as the host name is e.g. `web-EU-1.elastic.co` and
only the number changes non-randomly we deem this change ok)
* The `offset` will change be more realistic now: Before it changed randomly and
now the offset increases by the average event size up to a certain maximum.
* The current `@timestamp` will be retrieved only once per bulk. For documents
within a bulk we'll advance the microsecond portion by `1 / bulk size`
microseconds.

We've measured the performance impact of this change by stubbing out
Elasticsearch with nginx and running the `index-logs-fixed-daily-volume`
challenge with the following track parameters:

* `bulk_size`: 20000
* `bulk_indexing_clients`: 16
* `number_of_days`: 1
* `daily_logging_volume`: "20GB"

We have measured the following median indexing throughput in our test
environment:

* baseline (master): 153476 docs/s
* Using a deterministic `hostname` and `offset`: 174371 docs/s
* All three measures together: 222611 docs/s

This means we improve the maximum achievable indexing throughput by 45% in this
configuration.
cbuescher pushed a commit to cbuescher/elasticsearch that referenced this issue Oct 2, 2023
@danielmitterdorfer
Update Rally config file template
272f693 
With this commit we update the Rally config file template so it does not
include any JDK references. With elastic/rally#518 we will resolve JDK
versions using the `JAVAx_HOME` environment variables.

Relates elastic/rally#518
Closes elastic#70
Relates elastic#71
This issue was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
>enhancement v0.06.0
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kimchy