Skip to content
This repository was archived by the owner on May 14, 2022. It is now read-only.

Experiment with nailgun #458

Closed
escowles opened this issue Mar 3, 2016 · 3 comments
Closed

Experiment with nailgun #458

escowles opened this issue Mar 3, 2016 · 3 comments

Comments

@escowles
Copy link
Member

escowles commented Mar 3, 2016

No description provided.

@escowles
Copy link
Member Author

escowles commented Mar 3, 2016

Nailgun creates a daemon which preserves the JVM state across multiple executions, in theory reducing the overhead associated with the JVM startup. See samvera/hydra-file_characterization#21 for more info on using Nailgun for improving FITS performance.

@escowles
Copy link
Member Author

escowles commented Mar 3, 2016

Running FITS with Nailgun, on a sample dataset of 98 TIFFs (~100MB each), was significantly faster than running it without Nailgun. Normal execution took 479 sec, Nailgun execution took 296.

@escowles escowles closed this as completed Mar 7, 2016
@grosscol
Copy link

What happens with the memory footprint? I recall having a slow memory leak when I was using Nailgun with JRuby on Java7 VMs.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@escowles @grosscol