Tweak for Low-Memory usage. Make compatible with raspberry pi. #2
Conversation
This file contains 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
The orientDB tries to allocate too much memory for a small raspberry pi.
This commit reduces the RAM usage of the internally used orientdb.
Important: Start the docker container with
--env ORIENTDB_OPTS_MEMORY="-Xms200M -Xmx200M -XX:MaxDirectMemorySize=200M"
to further reduce memory usage.
The settings for the orientdb have been found by try and error.
This is a list of all [db parameters](http://www.orientdb.com/docs/last/admin/Configuration.html).
I used some suggestions from
[this](https://stackoverflow.com/questions/37016787/orientdb-2-1-11-java-process-consuming-too-much-memory/46023793)
stack overflow question and just went through all the config parameters of the documentation
and tweaked those that sounded like they would affect memory consumption.
The original base image did not have any binaries for the arm32 architecture on docker hub.
In fact, I did not found a single arm32 image with openjdk 12 jre/jdk. (why?)
So I tried openjdk 14 and it seems to work.
The new image is larger and based on ubuntu, but that's better than nothing.
In the orientdb server.sh there is this piece of code:
# Raspberry Pi check (Java VM does not run with -server argument on ARMv6)
if [ `uname -m` != "armv6l" ]; then
JAVA_OPTS="$JAVA_OPTS -server "
fi
I don't know if this workaround is still needed for the architecture `armv7l` of
the newer Raspberry Pis, but I simply replaced armv6l by armv7l. Seems to work.
|
Thanks for doing this legwork and sharing! Would you mind sharing a bit about your use case? After making these changes, how many simultaneous users do you need to support? How large are your models? Which features of Convergence are you using (e.g. the modules at the right in our API docs)? I agree that there are likely many smaller deployments of Convergence that could benefit from this smaller memory footprint. |
|
For our current use case, ~5 Users, ~10000 text documents with 1 to 60 KB, on average 10KB. It's an internal tool to create test scripts.
The plan is to use Activities and Real time data (only text and a flag that editing is locked / the document has been superseded). Users are authenticated through JWT. The actual data truth will be outside of converge. (The user can create checkpoints / commit the current document as new numbered revision)
During testing, convergence was sometimes unresponsive. So the current plan is to have a backup solution when the convergence server is down. (via a kind of rebase and manual conflict resolution like in git)
Am 8. Oktober 2020 00:48:20 MESZ schrieb Alec LaLonde <[email protected]>:
…Thanks for doing this legwork and sharing! Would you mind sharing a
bit about your use case? After making these changes, how many
simultaneous users do you need to support? How large are your models?
Which features of Convergence are you using (e.g. the modules at the
right in our [API docs](https://docs.convergence.io/js-api/))?
I agree that there are likely many smaller deployments of Convergence
that could benefit from this smaller memory footprint.
--
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub:
#2 (comment)
--
Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.
|
|
Wanted to check back in and see if you had any progress. We are re-tooling our containers and I am wondering if this configuration worked for you in the long run. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Hello Convergence Team
The current convergence-omnibus image is very memory hungry. It is almost impossible to run more than this application on a less powerful machine.
This PR is not intended for merging but as a proposal for an additional, less hungry docker image.
Maybe you could also add a third alternative container that can be used on a raspberry pi.
Please see the commit message for details.
Best Regards, Daniel
(See also https://github.com/Phaiax/django-convergence-ace/ )