May 2009 - infinispan-issues - Jboss List Archives

[JBoss JIRA] Resolved: (ISPN-49) Perf test using JBoss Marshalling marshaller layer

by Galder Zamarreño (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-49?page=com.atlassian.jira.plugin... ] Galder Zamarreño resolved ISPN-49. ---------------------------------- Fix Version/s: (was: 4.0.0.GA) Resolution: Done Happy with performance related to JBMAR solution after several tweaks in JBMAR and the implementation itself. > Perf test using JBoss Marshalling marshaller layer > -------------------------------------------------- > > Key: ISPN-49 > URL: https://jira.jboss.org/jira/browse/ISPN-49 > Project: Infinispan > Issue Type: Task > Components: RPC > Reporter: Galder Zamarreño > Assignee: Galder Zamarreño > Fix For: 4.0.0.BETA1 > > Attachments: round-lab-0.zip, round-lab-1.zip, run1.zip, run2.zip, run3.zip, run4.zip > > > Run performance comparisons between Infinispan using home grown marshalling framework > and the one based on JBoss Marshalling. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Resolved: (ISPN-42) Object stream pooling

by Galder Zamarreño (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-42?page=com.atlassian.jira.plugin... ] Galder Zamarreño resolved ISPN-42. ---------------------------------- Resolution: Done > Object stream pooling > --------------------- > > Key: ISPN-42 > URL: https://jira.jboss.org/jira/browse/ISPN-42 > Project: Infinispan > Issue Type: Feature Request > Components: RPC > Reporter: Manik Surtani > Assignee: Galder Zamarreño > Fix For: 4.0.0.BETA1, 4.0.0.GA > > > Make sure we pool OIS and OOS when marshalling/unmarshalling state as this is expensive to create. Consider XNIO for the necessary building blocks. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-25) JGroups ships with an XSD schema. Use this schema when validating configurations.

by Vladimir Blagojevic (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-25?page=com.atlassian.jira.plugin... ] Vladimir Blagojevic commented on ISPN-25: ----------------------------------------- We cannot even start this issue until JGroups 2.8 is used as transport. > JGroups ships with an XSD schema. Use this schema when validating configurations. > ---------------------------------------------------------------------------------- > > Key: ISPN-25 > URL: https://jira.jboss.org/jira/browse/ISPN-25 > Project: Infinispan > Issue Type: Feature Request > Components: Configuration, RPC > Reporter: Manik Surtani > Assignee: Vladimir Blagojevic > Fix For: 4.0.0.BETA1, 4.0.0.GA > > -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-46) Feature comparison with other products

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-46?page=com.atlassian.jira.plugin... ] Manik Surtani commented on ISPN-46: ----------------------------------- Should be in the form of a feature matrix on our public wiki. Other products to compare are EHCache, IBM's WebSphere eXtreme Scale and GemStone's GemFire. > Feature comparison with other products > -------------------------------------- > > Key: ISPN-46 > URL: https://jira.jboss.org/jira/browse/ISPN-46 > Project: Infinispan > Issue Type: Task > Reporter: Mircea Markus > Assignee: Manik Surtani > Fix For: 4.0.0.BETA1, 4.0.0.GA > > > Compare (feature wise) against other products: > - Coherence, Terracotta and gigaspaces, hadoop, mogileFS -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Updated: (ISPN-49) Perf test using JBoss Marshalling marshaller layer

by Galder Zamarreño (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-49?page=com.atlassian.jira.plugin... ] Galder Zamarreño updated ISPN-49: --------------------------------- Attachment: run4.zip Here's some graphs after implementing a solution based purely on ObjectTable and pooling marshaller with thread locals. Performance overall looks much better but I'm intrigued about 4-6 node async repl perf with JBMAR. This is currently being investigated. > Perf test using JBoss Marshalling marshaller layer > -------------------------------------------------- > > Key: ISPN-49 > URL: https://jira.jboss.org/jira/browse/ISPN-49 > Project: Infinispan > Issue Type: Task > Components: RPC > Reporter: Galder Zamarreño > Assignee: Galder Zamarreño > Fix For: 4.0.0.BETA1, 4.0.0.GA > > Attachments: round-lab-0.zip, round-lab-1.zip, run1.zip, run2.zip, run3.zip, run4.zip > > > Run performance comparisons between Infinispan using home grown marshalling framework > and the one based on JBoss Marshalling. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-70) Transparent eager locking for transactions

by Vladimir Blagojevic (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-70?page=com.atlassian.jira.plugin... ] Vladimir Blagojevic commented on ISPN-70: ----------------------------------------- A new <transaction> element attribute "useEagerLocking" has been added. By default it is false. If true, implicit eager locking is turned on for transactions. > Transparent eager locking for transactions > ------------------------------------------ > > Key: ISPN-70 > URL: https://jira.jboss.org/jira/browse/ISPN-70 > Project: Infinispan > Issue Type: Feature Request > Components: Locking and Concurrency, Transactions > Reporter: Manik Surtani > Assignee: Vladimir Blagojevic > Fix For: 4.0.0.BETA1, 4.0.0.GA > > > Allow a configuration attribute on <transaction /> to be able to specify whether cluster-wide locks are acquired eagerly or lazily. The current scheme (lazy) should be the default. > This could be implemented by broadcasting LockControlCommands alongside acquiring local locks for necessary keys. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Resolved: (ISPN-70) Transparent eager locking for transactions

by Vladimir Blagojevic (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-70?page=com.atlassian.jira.plugin... ] Vladimir Blagojevic resolved ISPN-70. ------------------------------------- Resolution: Done We agreed that the optimal solution, taking all trade offs into account, is to create a new interceptor. It just gets too messy if we subclass TxInterceptor since we have to then handle DistTxInterceptor and adding this functionality to LockingInterceptor blows it up quite a bit. Existing LockIterceptor#visitLockControlCommand (implemented for ISPN-48) has been augmented slightly to handle specific cases of implicit locking. A new interceptor was created - ImplicitEagerLockingInterceptor > Transparent eager locking for transactions > ------------------------------------------ > > Key: ISPN-70 > URL: https://jira.jboss.org/jira/browse/ISPN-70 > Project: Infinispan > Issue Type: Feature Request > Components: Locking and Concurrency, Transactions > Reporter: Manik Surtani > Assignee: Vladimir Blagojevic > Fix For: 4.0.0.BETA1, 4.0.0.GA > > > Allow a configuration attribute on <transaction /> to be able to specify whether cluster-wide locks are acquired eagerly or lazily. The current scheme (lazy) should be the default. > This could be implemented by broadcasting LockControlCommands alongside acquiring local locks for necessary keys. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-70) Transparent eager locking for transactions

by Vladimir Blagojevic (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-70?page=com.atlassian.jira.plugin... ] Vladimir Blagojevic commented on ISPN-70: ----------------------------------------- Unit test is SyncReplImplicitLockingTest > Transparent eager locking for transactions > ------------------------------------------ > > Key: ISPN-70 > URL: https://jira.jboss.org/jira/browse/ISPN-70 > Project: Infinispan > Issue Type: Feature Request > Components: Locking and Concurrency, Transactions > Reporter: Manik Surtani > Assignee: Vladimir Blagojevic > Fix For: 4.0.0.BETA1, 4.0.0.GA > > > Allow a configuration attribute on <transaction /> to be able to specify whether cluster-wide locks are acquired eagerly or lazily. The current scheme (lazy) should be the default. > This could be implemented by broadcasting LockControlCommands alongside acquiring local locks for necessary keys. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-32) Integrate Query API

by Igor Lobanov (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-32?page=com.atlassian.jira.plugin... ] Igor Lobanov commented on ISPN-32: ---------------------------------- Please, consider implementing JavaSpaces API for querying IMDG. > Integrate Query API > ------------------- > > Key: ISPN-32 > URL: https://jira.jboss.org/jira/browse/ISPN-32 > Project: Infinispan > Issue Type: Feature Request > Components: Querying > Reporter: Manik Surtani > Assignee: Manik Surtani > Fix For: 5.1.0.BETA1, 5.1.0.GA > > > Grab this from JBCS -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

[JBoss JIRA] Commented: (ISPN-59) Add unit test that verifies number of written bytes

by Galder Zamarreño (JIRA)

[ https://jira.jboss.org/jira/browse/ISPN-59?page=com.atlassian.jira.plugin... ] Galder Zamarreño commented on ISPN-59: -------------------------------------- Setting version to 0 reduces the payload avoiding block mode (each object in a block) but could potentially be a security issue and could spoil things when trying to serialize a spec-compliant Serializable object which relies on reading a -1 to know when its data ends will "overrun" the buffer if there's no block mode in place. This is too bigger risk so won't be using version 0 for the moment. > Add unit test that verifies number of written bytes > --------------------------------------------------- > > Key: ISPN-59 > URL: https://jira.jboss.org/jira/browse/ISPN-59 > Project: Infinispan > Issue Type: Sub-task > Components: RPC > Reporter: Galder Zamarreño > Assignee: Galder Zamarreño > Fix For: 4.0.0.BETA1, 4.0.0.GA > > Attachments: MarshallerTest.java, org.infinispan.marshall.MarshallersTest-v2-onlyobjecttable.xml, org.infinispan.marshall.MarshallersTest-v2-onlyobjecttable_and_version=0.xml, org.infinispan.marshall.MarshallerTest.xml > > > Add written byte number comparisons to present JBossMarshallerTest to make sure that it writes > less bytes that existing MarshallerImpl and standard Java Serialization (ObjectStreamMarshaller) -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

15 years, 7 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

infinispan-issues May 2009