[JBoss JIRA] (ARTIF-729) Enable query caching and hibernate-infinispan 2LC by default
by Brett Meyer (JIRA)
Brett Meyer created ARTIF-729:
---------------------------------
Summary: Enable query caching and hibernate-infinispan 2LC by default
Key: ARTIF-729
URL: https://issues.jboss.org/browse/ARTIF-729
Project: Artificer
Issue Type: Feature Request
Reporter: Brett Meyer
Assignee: Brett Meyer
Fix For: 1.0.0.Beta2
Profiling has shown that repeated queries are a huge CPU hot spot. Example: when uploading the Artificer distro ZIP itself, the HibernateRelationshipFactory makes repeated queries (by UUID) to set up relationship targets. Since many of them are repeats (ex: find the parent distro artifact), the query cache and 2LC would be immensely helpful.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
9 years, 6 months
[JBoss JIRA] (ARTIF-728) Add upload endpoint that takes a local file path
by Brett Meyer (JIRA)
Brett Meyer created ARTIF-728:
---------------------------------
Summary: Add upload endpoint that takes a local file path
Key: ARTIF-728
URL: https://issues.jboss.org/browse/ARTIF-728
Project: Artificer
Issue Type: Feature Request
Reporter: Brett Meyer
Assignee: Brett Meyer
On the server and all clients, add a new upload endpoint that takes a local file path, instead of an InputStream/byte[]. If the file is available on/to the server's machine, there should be no reason why we can't simply use it, rather than pass it through the web service.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
9 years, 6 months
[JBoss JIRA] (ARTIF-727) Add non-blocking upload endpoints
by Brett Meyer (JIRA)
Brett Meyer created ARTIF-727:
---------------------------------
Summary: Add non-blocking upload endpoints
Key: ARTIF-727
URL: https://issues.jboss.org/browse/ARTIF-727
Project: Artificer
Issue Type: Feature Request
Reporter: Brett Meyer
Assignee: Brett Meyer
Uploading large archives can take a while, due to expansion, processing, and derivation. For instance, uploading the entire Artificer distro ZIP takes around 4 minutes. While that's reasonable (at least in my opinion), all of our upload endpoints are currently blocking.
Blocking may be desirable for some client apps, so definitely don't remove that. But, adding additional, non-blocking methods to the web service, java client, cli, web browser, and ejb would be helpful. The non-blocking method should at least block until the primary artifact is persisted, returning its metadata. But the rest should be async.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
9 years, 6 months
[JBoss JIRA] (ARTIF-498) Provide a mechanism to consume "seed data"
by Brett Meyer (JIRA)
[ https://issues.jboss.org/browse/ARTIF-498?page=com.atlassian.jira.plugin.... ]
Brett Meyer closed ARTIF-498.
-----------------------------
Resolution: Won't Fix
> Provide a mechanism to consume "seed data"
> ------------------------------------------
>
> Key: ARTIF-498
> URL: https://issues.jboss.org/browse/ARTIF-498
> Project: Artificer
> Issue Type: Enhancement
> Reporter: Eric Wittmann
> Assignee: Brett Meyer
>
> This is a feature request from David based on his testing of 1.3.0.Beta1 of dtgov:
> "Would be very nice for the user, that the first time dtgov is executed, the cli-commands file is loaded and all the dtgov artifacts are installed in s-ramp."
> I think the idea is that currently the user must do this manually after installing and launching s-ramp/dtgov. Instead, it would be interesting to have a way to consume this data on startup. I'm adding this as a suggested enhancment to s-ramp. If we decide this feature is not appropriate for s-ramp, then we should move it to a DTGov feature request and consider it in that context.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
9 years, 6 months