db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "sagar (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DERBY-6809) Java 1.8 feature use
Date Tue, 05 Dec 2017 05:46:00 GMT

    [ https://issues.apache.org/jira/browse/DERBY-6809?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16278051#comment-16278051

sagar commented on DERBY-6809:

Hi Rick 

Two questions.

I checked out 10.14 as its on Java 1.8 which I have right now. Dunno if and when I would want
to be a committer. Still early days. I could make few changes to small part and then commit
in the branch possibly which someone could have a look at and then pull it up to the main

Any other direction suggestion for me would be welcome. Better have someone throw light than

Secondly as you can see, I have got started somewhere.

So, if Vectors were replaced then they either missed out a few or the commit wasnt merged.

But my question was specifically to the org.apache.derby.impl.store.access.sort.MergeInserter
class and if it was safe to replace Vectors with unsynchronized Collections. The reason for
the question is that since Vectors were used it means the Author wanted synchronized gurantee.

And if thats the case then we can use modern optimized Concurrency enabled Collections.

Also, I will evaluate each and every change as small as this (using a different type) for
performance scalability balance and then make decisions, is what I have thought.

I also understand that just because there is C programming language,  I shouldnt go about
replacing Assembly Code blindly. 
Thats the base direction I have kept for myself.

But also, I have assumed that the JAVA COMPILER and JVM will get smarter and optimized in
the future and hence something of an overhead today will get auto optimized by the future

for eg. LAMBDAS vs AIC ... Depends on scenarios but the hope of future enhancements in LAMBDAS
makes me replace AIC with LAMBDAS. Also, since derby is a server or a long running process,
I feel that out of the box LAMBDAS will make DERBY faster than AIC, as with frequent calls
LAMBDAS will perform better for a long running process.

Currently, my approach is purely horizontal to check if there  can be an easy low hanging
fruit to be consumed without too much effort but by making a full pass to each and every file.

So awaiting direction on which codebase to start on. 


> Java 1.8 feature use
> --------------------
>                 Key: DERBY-6809
>                 URL: https://issues.apache.org/jira/browse/DERBY-6809
>             Project: Derby
>          Issue Type: Improvement
>          Components: Network Server
>    Affects Versions:
>            Reporter: sagar
>         Attachments: 2017-12-04-143613_1366x768_scrot.png
> Suggestion ...
> Is it possible to auto modify the existing source code using tools like Netbeans, and
take advantage of the new features in JDK 1.8 for better multiuser performance and better
utilization of current day multicore processors?
> Plainly put, can we have from 11.0 onwards a version of derby which takes advantage of
the advancements and new features in java 1.8 ... 

This message was sent by Atlassian JIRA

View raw message