Re: [java] JProb
> > Arnaud, > > Have you used JProbe much? Is it do anything better than YourKit/JProfiler? > > It does seem to be much more focused on J2EE profiling rather than > standalone apps. The lack of immediate integration with > eclipse/intelij has stopped me spending any more time looking at it. > > Martin That's the one I have used the most (for this project and prior to it). I like the way call graphs are displayed and it's easy to navigate (focus, prune, etc.). I have now been using JProfiler for several weeks and I would say that it does the job. Jprob is certainly more mature, it offers a "Race condition detection" feature that I don't think Jprofiler has. It also provides "Code coverage" but we can use Intellij for that. I also like the snapshot approach of Jprob even if this is a pain for CPU profiling. Regarding J2EE apps I have mainly used Optimizeit for that so I cannot say if Jprob is good at it. Arnaud
Re: [java] JProb
On 04/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > On Feb 4, 2008 11:52 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > On Mon, 2008-02-04 at 11:44 +, Aidan Skinner wrote: > > > On Feb 4, 2008 10:09 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > > > > > I also agree and I have told Quest Software that we would only use the > > > > product based on an offer similar to the JMeter one. > > > > > > What did they offer JMeter? > > > > > > - Aidan > > > > Sorry, Jprofiler (JMeter is free anyway) > > Ah, ok. I did wonder, thought you maybe meant they'd offered JMeter > licensces under different terms. :) > > - Aidan > -- > aim/y!:aidans42 g:[EMAIL PROTECTED] > http://aidan.skinner.me.uk/ > "Almost everything is imitation... The most original writers borrowed > from one another." - Voltaire Arnaud, Have you used JProbe much? Is it do anything better than YourKit/JProfiler? It does seem to be much more focused on J2EE profiling rather than standalone apps. The lack of immediate integration with eclipse/intelij has stopped me spending any more time looking at it. Martin -- Martin Ritchie
Re: [java] JProb
On Feb 4, 2008 11:52 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > On Mon, 2008-02-04 at 11:44 +, Aidan Skinner wrote: > > On Feb 4, 2008 10:09 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > > > I also agree and I have told Quest Software that we would only use the > > > product based on an offer similar to the JMeter one. > > > > What did they offer JMeter? > > > > - Aidan > > Sorry, Jprofiler (JMeter is free anyway) Ah, ok. I did wonder, thought you maybe meant they'd offered JMeter licensces under different terms. :) - Aidan -- aim/y!:aidans42 g:[EMAIL PROTECTED] http://aidan.skinner.me.uk/ "Almost everything is imitation... The most original writers borrowed from one another." - Voltaire
Re: [java] JProb
On Mon, 2008-02-04 at 11:44 +, Aidan Skinner wrote: > On Feb 4, 2008 10:09 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > I also agree and I have told Quest Software that we would only use the > > product based on an offer similar to the JMeter one. > > What did they offer JMeter? > > - Aidan Sorry, Jprofiler (JMeter is free anyway) Arnaud
Re: [java] JProb
On Feb 4, 2008 10:09 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > I also agree and I have told Quest Software that we would only use the > product based on an offer similar to the JMeter one. What did they offer JMeter? - Aidan -- aim/y!:aidans42 g:[EMAIL PROTECTED] http://aidan.skinner.me.uk/ "Almost everything is imitation... The most original writers borrowed from one another." - Voltaire
Re: [java] JProb
I also agree and I have told Quest Software that we would only use the product based on an offer similar to the JMeter one. Arnaud On Fri, 2008-02-01 at 17:35 +, Robert Godfrey wrote: > Definitely agree, and I don't think we should be endorsing one product > over another. Any company which is generous enough to offer licenses > in exchange merely for recognition of the donation I am very happy to > accept. Other restrictions or requests for publicity I think should > declined. > > -- Rob > > On 01/02/2008, Robert Greig <[EMAIL PROTECTED]> wrote: > > On 01/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > > > > > 'Recommended' is substantially stronger than the 'used by' endorsement > > > we give to other donors, I'm not entirely sure we should be showing > > > favouritism in that way. > > > > I would agree. Unless we actually *genuinely* think that it offers > > something significantly better than we get either with YourKit or the > > free-to-everyone Netbeans Profiler I don't think we should take up > > such a restrictive offer. > > > > At the very least I would expect them to offer licences to every Qpid > > committer. > > > > RG > >
Re: [java] JProb
Before you go too far down the track, this proposal needs to be brought to the attention of the incubator pmc who will give you advice. Craig On Feb 1, 2008, at 12:25 AM, Arnaud Simon wrote: Hello, I have been talking with the JProb people and even if they are not very much open source minded they agree to provide us with 5 free licenses. However, they will give us those licenses only under those conditions: 1. Feature JProbe as the recommended profiling tool to the Apache Qpid project community, including but not limited to a link and banner on the acknowledgement page of the project site (http://cwiki.apache.org/qpid/acknowledgement.html). 2. To recommend JProbe as the preferred profiler of the Apache Qpid project through a joint case study. Quest shall retain distribution rights of the article to be used in its marketing communications. 3. To participate in joint marketing activities including one joint Webcast on how JProbe is being used to solve performance issues. Quest shall retain distribution rights of the article to be used in its marketing communications. We already know that point 1. is acceptable. What do you think about 2 and 3? They want use to co-author a case study (one of us is ok) and they also want one of us to participate to one webcast they'll organize end of this year. In exchange they would provide us 5 node-locked licenses of JProbe with 1 year updates/support for free. If we decide to accept this offer we will need to provide them with 5 manes and the persons will need to fill and sign a license agreement. Please let me know what you think about that. Arnaud Craig Russell Architect, Sun Java Enterprise System http://java.sun.com/products/jdo 408 276-5638 mailto:[EMAIL PROTECTED] P.S. A good JDO? O, Gasp! smime.p7s Description: S/MIME cryptographic signature
Re: [java] JProb
Definitely agree, and I don't think we should be endorsing one product over another. Any company which is generous enough to offer licenses in exchange merely for recognition of the donation I am very happy to accept. Other restrictions or requests for publicity I think should declined. -- Rob On 01/02/2008, Robert Greig <[EMAIL PROTECTED]> wrote: > On 01/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > > > 'Recommended' is substantially stronger than the 'used by' endorsement > > we give to other donors, I'm not entirely sure we should be showing > > favouritism in that way. > > I would agree. Unless we actually *genuinely* think that it offers > something significantly better than we get either with YourKit or the > free-to-everyone Netbeans Profiler I don't think we should take up > such a restrictive offer. > > At the very least I would expect them to offer licences to every Qpid > committer. > > RG >
Re: [java] JProb
+1 On 01/02/2008, Rajith Attapattu <[EMAIL PROTECTED]> wrote: > > JProfiler offered us all licenses and only asked for a simple link. > I think it is unfair if we name JProbe the prefered profiler when they > only > offer 8 licenses. > > So I agree with Aidan and Rob on this. > > Rajith > > On Feb 1, 2008 12:21 PM, Robert Greig <[EMAIL PROTECTED]> wrote: > > > On 01/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > > > > > 'Recommended' is substantially stronger than the 'used by' endorsement > > > we give to other donors, I'm not entirely sure we should be showing > > > favouritism in that way. > > > > I would agree. Unless we actually *genuinely* think that it offers > > something significantly better than we get either with YourKit or the > > free-to-everyone Netbeans Profiler I don't think we should take up > > such a restrictive offer. > > > > At the very least I would expect them to offer licences to every Qpid > > committer. > > > > RG > > > > > > -- > Regards, > > Rajith Attapattu > Red Hat > blog: http://rajith.2rlabs.com/ >
Re: [java] JProb
JProfiler offered us all licenses and only asked for a simple link. I think it is unfair if we name JProbe the prefered profiler when they only offer 8 licenses. So I agree with Aidan and Rob on this. Rajith On Feb 1, 2008 12:21 PM, Robert Greig <[EMAIL PROTECTED]> wrote: > On 01/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > > > 'Recommended' is substantially stronger than the 'used by' endorsement > > we give to other donors, I'm not entirely sure we should be showing > > favouritism in that way. > > I would agree. Unless we actually *genuinely* think that it offers > something significantly better than we get either with YourKit or the > free-to-everyone Netbeans Profiler I don't think we should take up > such a restrictive offer. > > At the very least I would expect them to offer licences to every Qpid > committer. > > RG > -- Regards, Rajith Attapattu Red Hat blog: http://rajith.2rlabs.com/
Re: [java] JProb
On 01/02/2008, Aidan Skinner <[EMAIL PROTECTED]> wrote: > 'Recommended' is substantially stronger than the 'used by' endorsement > we give to other donors, I'm not entirely sure we should be showing > favouritism in that way. I would agree. Unless we actually *genuinely* think that it offers something significantly better than we get either with YourKit or the free-to-everyone Netbeans Profiler I don't think we should take up such a restrictive offer. At the very least I would expect them to offer licences to every Qpid committer. RG
Re: [java] JProb
On Feb 1, 2008 8:25 AM, Arnaud Simon <[EMAIL PROTECTED]> wrote: > 1. Feature JProbe as the recommended profiling tool to the Apache > Qpid project community, including but not limited to a link and banner > on the acknowledgement page of the project site > (http://cwiki.apache.org/qpid/acknowledgement.html). 'Recommended' is substantially stronger than the 'used by' endorsement we give to other donors, I'm not entirely sure we should be showing favouritism in that way. - Aidan -- aim/y!:aidans42 g:[EMAIL PROTECTED] http://aidan.skinner.me.uk/ "Almost everything is imitation... The most original writers borrowed from one another." - Voltaire
Re: [java] JProb
On 01/02/2008, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > On Fri, 2008-02-01 at 09:05 +, Martin Ritchie wrote: > > On 01/02/2008, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > > > > > > > > > Just my opinion, but I think that the combination of recommending JProbe > > > > as the 'preferred profiler' while only allowing 5 members of the project > > > >doesn't seem right. Using the project as a whole to get free licenses > > > > (of non-free software) for some seems to somehow detract from the sense > > > > of a united community. > > > > > > They are saying 5 as this is the number I had in mind when asked how > > > many programmers will need a license. We can ask them to unrestrict the > > > number of license and stands that any project member can request one. It > > > would definitely be more acceptable. > > > > > > Arnaud > > > > We should probably ask legal-discuss for guidance on points 2 & 3 as > > we would be marketing Apache as well as Apache Qpid. > > > > What does JProbe offer that JProfiler and YourKit don't? How many > > profilers do we need and how will saying that JProbe is the profiler > > of choice sit when we have acknowledgements to JProfiler and YourKit > > sitting along side on the acknowledgement page. > > That's a fair point. I personally was considering JProbe as I already > knew the product. I should confess that I haven't made a comparative > study. I got this offer from JProbe that we can obviously decline. > However, even if my current feeling is that JProfiler has all the > required features I would vote +1 for JProbe only because of laziness. > This being said I can of think that they are asking far too much. As I > said those guys are not open source minded. Perhaps we can change their minds. I've been using YourKit and JProfiler over the last couple of days trying to get to the bottom of a problem... I might as well try JProbe if they have a trial period. If they can help me sort the problem then I will be +1-ing too. :) > > Arnaud are you in the process of making the ack page to put JProfiler on it? > > I have added JProfiler on an ack page: > http://cwiki.apache.org/qpid/acknowledgment.html > > I have the feeling that we should also add IntelliJ as some of us are > using a free license. Ah indeedy. I'd even forgotten I had the free license for 7. My paid up license worked in 5 and 6. -- Martin Ritchie
Re: [java] JProb
On Fri, 2008-02-01 at 09:05 +, Martin Ritchie wrote: > On 01/02/2008, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > > > > > > Just my opinion, but I think that the combination of recommending JProbe > > > as the 'preferred profiler' while only allowing 5 members of the project > > >doesn't seem right. Using the project as a whole to get free licenses > > > (of non-free software) for some seems to somehow detract from the sense > > > of a united community. > > > > They are saying 5 as this is the number I had in mind when asked how > > many programmers will need a license. We can ask them to unrestrict the > > number of license and stands that any project member can request one. It > > would definitely be more acceptable. > > > > Arnaud > > We should probably ask legal-discuss for guidance on points 2 & 3 as > we would be marketing Apache as well as Apache Qpid. > > What does JProbe offer that JProfiler and YourKit don't? How many > profilers do we need and how will saying that JProbe is the profiler > of choice sit when we have acknowledgements to JProfiler and YourKit > sitting along side on the acknowledgement page. That's a fair point. I personally was considering JProbe as I already knew the product. I should confess that I haven't made a comparative study. I got this offer from JProbe that we can obviously decline. However, even if my current feeling is that JProfiler has all the required features I would vote +1 for JProbe only because of laziness. This being said I can of think that they are asking far too much. As I said those guys are not open source minded. > Arnaud are you in the process of making the ack page to put JProfiler on it? I have added JProfiler on an ack page: http://cwiki.apache.org/qpid/acknowledgment.html I have the feeling that we should also add IntelliJ as some of us are using a free license.
Re: [java] JProb
On 01/02/2008, Arnaud Simon <[EMAIL PROTECTED]> wrote: > > > > > Just my opinion, but I think that the combination of recommending JProbe > > as the 'preferred profiler' while only allowing 5 members of the project > >doesn't seem right. Using the project as a whole to get free licenses > > (of non-free software) for some seems to somehow detract from the sense > > of a united community. > > They are saying 5 as this is the number I had in mind when asked how > many programmers will need a license. We can ask them to unrestrict the > number of license and stands that any project member can request one. It > would definitely be more acceptable. > > Arnaud We should probably ask legal-discuss for guidance on points 2 & 3 as we would be marketing Apache as well as Apache Qpid. What does JProbe offer that JProfiler and YourKit don't? How many profilers do we need and how will saying that JProbe is the profiler of choice sit when we have acknowledgements to JProfiler and YourKit sitting along side on the acknowledgement page. Arnaud are you in the process of making the ack page to put JProfiler on it? -- Martin Ritchie
Re: [java] JProb
> > Just my opinion, but I think that the combination of recommending JProbe > as the 'preferred profiler' while only allowing 5 members of the project >doesn't seem right. Using the project as a whole to get free licenses > (of non-free software) for some seems to somehow detract from the sense > of a united community. They are saying 5 as this is the number I had in mind when asked how many programmers will need a license. We can ask them to unrestrict the number of license and stands that any project member can request one. It would definitely be more acceptable. Arnaud
Re: [java] JProb
Arnaud Simon wrote: Hello, I have been talking with the JProb people and even if they are not very much open source minded they agree to provide us with 5 free licenses. However, they will give us those licenses only under those conditions: 1. Feature JProbe as the recommended profiling tool to the Apache Qpid project community, including but not limited to a link and banner on the acknowledgement page of the project site (http://cwiki.apache.org/qpid/acknowledgement.html). 2. To recommend JProbe as the preferred profiler of the Apache Qpid project through a joint case study. Quest shall retain distribution rights of the article to be used in its marketing communications. 3. To participate in joint marketing activities including one joint Webcast on how JProbe is being used to solve performance issues. Quest shall retain distribution rights of the article to be used in its marketing communications. We already know that point 1. is acceptable. What do you think about 2 and 3? They want use to co-author a case study (one of us is ok) and they also want one of us to participate to one webcast they'll organize end of this year. In exchange they would provide us 5 node-locked licenses of JProbe with 1 year updates/support for free. If we decide to accept this offer we will need to provide them with 5 manes and the persons will need to fill and sign a license agreement. Please let me know what you think about that. Just my opinion, but I think that the combination of recommending JProbe as the 'preferred profiler' while only allowing 5 members of the project doesn't seem right. Using the project as a whole to get free licenses (of non-free software) for some seems to somehow detract from the sense of a united community.
