December 2009 - jboss-dev-forums - Jboss List Archives

[JBoss ESB Development] - Re: Performance Improvement Proposal: Cache result of HttpSe

by Kevin.Conner＠jboss.com

"dward" wrote : I don't see how using ConcurrentHashMap would be any different than Collections.synchronizedMap(new HashMap()) In answer to this part, the difference is in the enforced synchronization for readers. With ConcurrentHashMap you can have multiple readers accessing the map concurrently. This may not be such a big difference on some machines but why have a synchronization point when you don't need it? Kev View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269527#4269527 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269527

16 years, 1 month

1
0
0 / 0

[JBoss ESB Development] - Re: Performance Improvement Proposal: Cache result of HttpSe

by Kevin.Conner＠jboss.com

Useful information for this discussion. http://www.cs.umd.edu/users/pugh/java/memoryModel/jsr-133-faq.html View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269526#4269526 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269526

16 years, 1 month

1
0
0 / 0

[JBoss Web Development] - Re: jbosssvc.exe on Windows 7

by mladen.turk＠jboss.com

Try running the cmd.exe as Administrator and then CD to the bin folder and execute 'service install' Note that if UAC is enabled you will have to do that even if logged in as Administrator. Also make sure the location where JBoss is installed has Full privileges enabled for Local Service account (or the one you later choose to use as service's account) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269525#4269525 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269525

16 years, 1 month

1
0
0 / 0

[JBoss ESB Development] - Re: Performance Improvement Proposal: Cache result of HttpSe

by Kevin.Conner＠jboss.com

"dward" wrote : I don't see how using ConcurrentHashMap would be any different than Collections.synchronizedMap(new HashMap()). So it would still open the door for multiple calls to request.getLocalName(). Wrong issue. "dward" wrote : I also don't want to use oswego classes if not necessary. Sorry, why do you need to use oswego? ConcurrentHashMap is in java.util.concurrent as of jdk1.5. Kev View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269523#4269523 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269523

16 years, 1 month

1
0
0 / 0

[JBoss ESB Development] - Re: Performance Improvement Proposal: Cache result of HttpSe

by Kevin.Conner＠jboss.com

"dward" wrote : That's good to know Bill, thanks. The difference with what I'm doing is that it's not a member variable (the map) that is being set lazily, but a populated entry in the map, so we just have to be careful against ConcurrentModificationExceptions. That is certainly one issue but nothing to do with the memory model. "dward" wrote : Kev, if you're okay with me not using ConcurrentHashMap (and instead using synchronized as I elaborated on above), let me know. Either way, I'll wait for your word before I commit to the performance workspace and 4.7 CP branch. Thanks. Your code above is not thread safe, please do not commit that. Double checking is not where it is falls down, rather it is the updates to the structure of the map that would not be propagated correctly in certain memory models. The only reason Bill's example works is because of the volatile keyword, as that now has the same semantics as synchronized, wrt memory. Kev View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269520#4269520 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269520

16 years, 1 month

1
0
0 / 0

[JBoss ESB Development] - Re: Performance Improvement Proposal: Cache result of HttpSe

by Kevin.Conner＠jboss.com

"bill.burke(a)jboss.com" wrote : FYI, doublecheck works: But that's not the issue here ;) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269519#4269519 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269519

16 years, 1 month

1
0
0 / 0

[JBoss Profiler Development] - Re: Beta6 remaining issues

by jaikiran

"kukeltje" wrote : | | btw, I saw you are also working on the JCA implementation of JBoss. What is the best forum to ask about a 'transactional file store' so it can kind of run in a transaction? It does not need to be fully 100% safe, just in a way that I do not have to program a lot myself. Just a FYI - http://www.jboss.org/jbosstm/fileio/. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269515#4269515 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269515

16 years, 1 month

1
0
0 / 0

[JBoss Web Services Development] - Re: JBossWS Deployers Architecture

by richard.opalka＠jboss.com

"bill.burke(a)jboss.com" wrote : | Are the aspectized deployers just a abstraction for JBossWS so that it can move/integration smoothly between different deployment mechanisms? | Exactly, U got it ;) Let me explain you our architecture more high level. There are three main JBossWS integration points: * SPI * ASIL (application server integration layer) * SSIL (SOAP stack integration layer) Our SPI provides abstraction for different JBossAS servers and SOAP stacks. The SPI help us to be both JBossAS and SOAP stack agnostic. To achieve it, our SPI have its own UMDM (universal meta data model) that is generated from both JBossAS and SOAP stack MD (meta data). The ASIL always resides in JBossAS/webservices module and implements our SPI (to be JBossAS agnostic). Of course it has no notion about SOAP stack and uses our SPI to integrate with it. The goals of ASIL are: * AS components integration (EJB3 invocations, injection, Web MD modification) * AS MD => UMDM translation (translate JBossAS MD to JBossAS agnostic MD) * deployers integration (moves deployers abstraction one step higher, we have custom WSDeploymentAspectDeployer that allows us to decouple from JBossAS specific deployers and thus allows us to support multible JBossAS versions as pictured above) The SSIL allows us to support different SOAP stacks. Currently we provide support for three SOAP stacks: * Native (SVN, Home Page) * Metro (SVN, Home Page) * CXF (SVN, Home Page) The SSIL always resides in separate JBoss SVN and implements our SPI (to be SOAP stack agnostic). Of course it has no notion about JBossAS server and uses our SPI to integrate with it. So back to answer your question. "bill.burke(a)jboss.com" wrote : | Are the aspectized deployers just a abstraction for JBossWS so that it can move/integration smoothly between different deployment mechanisms? | We have WSDeploymentAspectDeployer to decouple from JBossAS internals and allow us to execute both JBossAS and SOAP stack agnostic DAs (Deployment Aspects). View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269512#4269512 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269512

16 years, 1 month

1
0
0 / 0

[JBoss Web Development] - Re: jbosssvc.exe on Windows 7

by maffeis

I'm having the same problem. Any solutions? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269504#4269504 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269504

16 years, 1 month

1
0
0 / 0

[JBoss Web Services Development] - Re: JBossWS Deployers Architecture

by richard.opalka＠jboss.com

"bill.burke(a)jboss.com" wrote : | All and all, it looks like things would be pretty straighforward to create a RESTEasy integration. Its just a matter of writing descriptor and real deployers and then importing, modifying, and exporting the correct metadata. | Exactly. "bill.burke(a)jboss.com" wrote : | BTW, would be interesting to hear what you think of the JBoss 5 deployment framework. So far, it looks like it is solving what it was intended to do. AlesJ, Warjort and others involved did an excellent job IMO. What about JBossWS team, we like JBoss5 deployment framework and had no real issues with it ;) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4269502#4269502 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4269502

16 years, 1 month

1
0
0 / 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums December 2009