Skip to content
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.

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

Distributed Class loading (Feature Request) #7394

Closed
the100rabh opened this issue Jan 25, 2016 · 6 comments

Comments

Projects
None yet
7 participants
@the100rabh
Copy link

commented Jan 25, 2016

There should be this ability in Hazelcast to do distributed loading of classes so that jobs like MapReduce and Aggregate can be done more effectively.

Things like custom predicates will not work for aggregate if using through Hazelcast client as in #7386

Without this feature, map reduce and aggregate becomes highly restrictive and if we need to add a new predicate, it would mean restarting the entire cluster which is not desirable at all.

@serkan-ozal

This comment has been minimized.

Copy link
Contributor

commented Jan 25, 2016

Hi @the100rabh

This may take your interest https://github.com/serkan-ozal/hermgen

Regards.

@bwzhang2011

This comment has been minimized.

Copy link

commented Jan 25, 2016

@serkan-ozal, maybe it's better to make it accepted and merged into hazelcast master branch and make it to be the feature.

@metanet

This comment has been minimized.

Copy link
Contributor

commented Jan 25, 2016

@serkan-ozal such feature could be nice

@vertex-github

This comment has been minimized.

Copy link

commented Feb 3, 2016

+1 (although I recall seeing a similar issue logged so this could be a duplicate). This is one of the great features of GridGain. I took a look at the hermgen project - its along the lines, but Id take it further - just load the classes directly on the node being used for processing (no need to hunt around the cluster - the invoking code should supply them). Being able to load classes dynamically opens HZ up to be used for distributed unit testing which is great. Use cases: 1. Just run this class on a node in the cluster - I dont care where - just pick one and run. 2. Run this class on a specific node in the cluster. Both of the usage models follow the existing HZ task distribution mechanism - we're just missing the dynamic class loading.

@bwzhang2011

This comment has been minimized.

Copy link

commented May 13, 2016

@serkan-ozal, any update with such issue ? hope such feature could be provided officially.

@mdogan mdogan modified the milestones: 3.8, Backlog Dec 29, 2016

@mdogan

This comment has been minimized.

Copy link
Member

commented Dec 29, 2016

Fixed by #9331

@mdogan mdogan closed this Dec 29, 2016

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.