[
https://issues.apache.org/jira/browse/TINKERPOP-2441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17215458#comment-17215458
]
ASF GitHub Bot commented on TINKERPOP-2441:
-------------------------------------------
spmallette edited a comment on pull request #1344:
URL: https://github.com/apache/tinkerpop/pull/1344#issuecomment-710124396
I did a really simple test - just ran a script through the console and
flipped on a JFR:
```groovy
// 3.4.8
graphson = new GraphSONMessageSerializerV3d0()
cluster =
Cluster.build().serializer(graphson).maxContentLength(6553600).maxConnectionPoolSize(16).minConnectionPoolSize(16).create()
client = cluster.connect()
g = traversal().withRemote(DriverRemoteConnection.using(client))
(0..<1024).each { g.V().repeat(both()).times(10).emit().fold().toList();[]
};[]
graphbin = new GraphBinaryMessageSerializerV1()
cluster =
Cluster.build().serializer(graphbin).maxContentLength(6553600).maxConnectionPoolSize(16).minConnectionPoolSize(16).create()
client = cluster.connect()
g = traversal().withRemote(DriverRemoteConnection.using(client))
(0..<1024).each { g.V().repeat(both()).times(10).emit().fold().toList();[]
};[]
// 3.4.9
graphson = new GraphSONMessageSerializerV3d0()
cluster =
Cluster.build().serializer(graphson).maxConnectionPoolSize(16).minConnectionPoolSize(16).create()
client = cluster.connect()
g = traversal().withRemote(DriverRemoteConnection.using(client))
(0..<1024).each { g.V().repeat(both()).times(10).emit().fold().toList();[]
};[]
graphbin = new GraphBinaryMessageSerializerV1()
cluster =
Cluster.build().serializer(graphbin).maxConnectionPoolSize(16).minConnectionPoolSize(16).create()
client = cluster.connect()
g = traversal().withRemote(DriverRemoteConnection.using(client))
(0..<1024).each { g.V().repeat(both()).times(10).emit().fold().toList();[]
};[]
```
Note that with 3.4.9 we don't need to even increase the `maxContentLength`
which is nice because it means users will have less chance of hitting that
problem early. The results had me questioning whether or not I even did this
test right as there was so much less IO going on:
## 3.4.8 and GraphSON
## 3.4.9 and GraphSON
## 3.4.8 and GraphBinary
## 3.4.9 and GraphBinary
In this process, i did some testing of the version interoperability by
accident using 3.4.8 across 3.4.9 and it worked nicely. didn't try 3.4.9 to
3.4.8 but i presume that would be fine as well given the initial description on
this PR.
When this merges I can write something up about it and CTR it into the
documentation.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Add compression to WebSocket frames sent from client
> ----------------------------------------------------
>
> Key: TINKERPOP-2441
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2441
> Project: TinkerPop
> Issue Type: Improvement
> Components: dotnet, driver, javascript, python, server
> Affects Versions: 3.5.0, 3.4.8
> Reporter: Divij Vaidya
> Priority: Minor
>
> Add compression for WebSocket frames. The compression was standardized in
> [https://tools.ietf.org/html/rfc7692] and Netty provides out of the box
> implementation for it, hence Java is easy but we should implement it for all
> clients.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
