You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@giraph.apache.org by Roman Shaposhnik <rv...@apache.org> on 2012/10/23 01:28:45 UTC
Giraph 0.2.0
Hi!
As was discussed on GIRAPH-285 it seems to be right time
to think about the release. It seems like Avery has agreed
to be an official RM and I'm volunteering for an RM sidekick
role ;-)
My plan is pretty simple: I would like to start with JIRA triaging,
figure out which ones folks consider to be blockers for 0.2 and
then help Avery curate the release branch. I am also volunteering
to test RCs with Hadoop 2.0.2-alpha (as part of the Bigtop 0.5.0)
and also do some limited testing with Hadoop 1.1.0 and CDH3/4
If this sounds like a good plan, here's a couple of areas where
help would still be appreciated:
* if you can test RCs with different versions of Hadoop -- please
let me know. In fact, my most rigorous testing (to the extent
it can be called that way) will be focused on Hadoop 2.0.2-alpha
* JIRA triage -- we've got 17 issues tagged with 0.2
http://s.apache.org/qH
and 22 issues that are assigned to folks, but not to the
0.2 release:
http://s.apache.org/zE
If you are an issue owner please let me know whether yours
could be considered a blocker for 0.2
Thanks,
Roman.
Re: Giraph 0.2.0
Posted by Doron Cohen <cd...@gmail.com>.
Hi, any chance to include GIRAPH-42 (so sent-messages counter > 0)?
Thanks, Doron
On Sat, Oct 27, 2012 at 2:05 AM, Avery Ching <ac...@apache.org> wrote:
> I agree with this. No showstoppers. The next version will have a ton of
> improvements though. Would be really nice to have GIRAPH-388 in though as
> it has a large performance impact.
>
> Avery
>
>
> On 10/26/12 2:48 PM, Roman Shaposhnik wrote:
>
>> On Thu, Oct 25, 2012 at 9:22 PM, Eli Reisman <ap...@gmail.com>
>> wrote:
>>
>>> I can say from the list you have, GIRAPH-260 and GIRAPH-263 are not
>>> blockers as far as I can see. Some of the other JIRA issues on the list
>>> might be easily resolved one way or the other as well at this point. I
>>> think both can safely be resolved at this point (260 can be a "won't
>>> fix",
>>> and 263 "fixed" as far as i can tell?)
>>>
>> Can we make a call on them one way or another? That would be helpful.
>> Personally, I'm going to triage a couple of the ones I had in there:
>> https://issues.apache.org/**jira/browse/GIRAPH-198<https://issues.apache.org/jira/browse/GIRAPH-198>
>> https://issues.apache.org/**jira/browse/GIRAPH-199<https://issues.apache.org/jira/browse/GIRAPH-199>
>>
>> At this point it looks as though we don't really have any showstoppers
>> for 0.2. At this point I'd like to proceed with a branch creation, RCs
>> and testing via Bigtop and other means on my side of things. Will
>> report accordingly.
>>
>> Thanks,
>> Roman.
>>
>
>
Re: Giraph 0.2.0
Posted by Avery Ching <ac...@apache.org>.
I agree with this. No showstoppers. The next version will have a ton of
improvements though. Would be really nice to have GIRAPH-388 in though
as it has a large performance impact.
Avery
On 10/26/12 2:48 PM, Roman Shaposhnik wrote:
> On Thu, Oct 25, 2012 at 9:22 PM, Eli Reisman <ap...@gmail.com> wrote:
>> I can say from the list you have, GIRAPH-260 and GIRAPH-263 are not
>> blockers as far as I can see. Some of the other JIRA issues on the list
>> might be easily resolved one way or the other as well at this point. I
>> think both can safely be resolved at this point (260 can be a "won't fix",
>> and 263 "fixed" as far as i can tell?)
> Can we make a call on them one way or another? That would be helpful.
> Personally, I'm going to triage a couple of the ones I had in there:
> https://issues.apache.org/jira/browse/GIRAPH-198
> https://issues.apache.org/jira/browse/GIRAPH-199
>
> At this point it looks as though we don't really have any showstoppers
> for 0.2. At this point I'd like to proceed with a branch creation, RCs
> and testing via Bigtop and other means on my side of things. Will
> report accordingly.
>
> Thanks,
> Roman.
Re: Giraph 0.2.0
Posted by Eli Reisman <ap...@gmail.com>.
The object weight thing (I think GIRAPH-260) can be resolved as "won't fix"
for now, we ended up handling this issue a different way and can come back
to that idea later. At that point, the idea will be good but that
particular JIRA might not be up to date.
Hyunsik was handling the jenkins stuff for 263, that should be resolved one
way or another soon as well. There are a few more.
As far as a graph generator, we have GIRAPH-26 waiting on a change from
Colt math libraries to mahout's and it should be good to go. I was told it
worked well by the guy who wrote it, but he has been out of touch lately.
I will take another peek at the other outstanding issues when I have time
and hopefully can sit down and get a few resolved.
Thanks!
Eli
On Fri, Oct 26, 2012 at 2:48 PM, Roman Shaposhnik <rv...@apache.org> wrote:
> On Thu, Oct 25, 2012 at 9:22 PM, Eli Reisman <ap...@gmail.com>
> wrote:
> > I can say from the list you have, GIRAPH-260 and GIRAPH-263 are not
> > blockers as far as I can see. Some of the other JIRA issues on the list
> > might be easily resolved one way or the other as well at this point. I
> > think both can safely be resolved at this point (260 can be a "won't
> fix",
> > and 263 "fixed" as far as i can tell?)
>
> Can we make a call on them one way or another? That would be helpful.
> Personally, I'm going to triage a couple of the ones I had in there:
> https://issues.apache.org/jira/browse/GIRAPH-198
> https://issues.apache.org/jira/browse/GIRAPH-199
>
> At this point it looks as though we don't really have any showstoppers
> for 0.2. At this point I'd like to proceed with a branch creation, RCs
> and testing via Bigtop and other means on my side of things. Will
> report accordingly.
>
> Thanks,
> Roman.
>
Re: Giraph 0.2.0
Posted by Roman Shaposhnik <rv...@apache.org>.
On Thu, Oct 25, 2012 at 9:22 PM, Eli Reisman <ap...@gmail.com> wrote:
> I can say from the list you have, GIRAPH-260 and GIRAPH-263 are not
> blockers as far as I can see. Some of the other JIRA issues on the list
> might be easily resolved one way or the other as well at this point. I
> think both can safely be resolved at this point (260 can be a "won't fix",
> and 263 "fixed" as far as i can tell?)
Can we make a call on them one way or another? That would be helpful.
Personally, I'm going to triage a couple of the ones I had in there:
https://issues.apache.org/jira/browse/GIRAPH-198
https://issues.apache.org/jira/browse/GIRAPH-199
At this point it looks as though we don't really have any showstoppers
for 0.2. At this point I'd like to proceed with a branch creation, RCs
and testing via Bigtop and other means on my side of things. Will
report accordingly.
Thanks,
Roman.
Re: Giraph 0.2.0
Posted by Eli Reisman <ap...@gmail.com>.
I can say from the list you have, GIRAPH-260 and GIRAPH-263 are not
blockers as far as I can see. Some of the other JIRA issues on the list
might be easily resolved one way or the other as well at this point. I
think both can safely be resolved at this point (260 can be a "won't fix",
and 263 "fixed" as far as i can tell?)
On Mon, Oct 22, 2012 at 4:28 PM, Roman Shaposhnik <rv...@apache.org> wrote:
> Hi!
>
> As was discussed on GIRAPH-285 it seems to be right time
> to think about the release. It seems like Avery has agreed
> to be an official RM and I'm volunteering for an RM sidekick
> role ;-)
>
> My plan is pretty simple: I would like to start with JIRA triaging,
> figure out which ones folks consider to be blockers for 0.2 and
> then help Avery curate the release branch. I am also volunteering
> to test RCs with Hadoop 2.0.2-alpha (as part of the Bigtop 0.5.0)
> and also do some limited testing with Hadoop 1.1.0 and CDH3/4
>
> If this sounds like a good plan, here's a couple of areas where
> help would still be appreciated:
> * if you can test RCs with different versions of Hadoop -- please
> let me know. In fact, my most rigorous testing (to the extent
> it can be called that way) will be focused on Hadoop 2.0.2-alpha
>
> * JIRA triage -- we've got 17 issues tagged with 0.2
> http://s.apache.org/qH
> and 22 issues that are assigned to folks, but not to the
> 0.2 release:
> http://s.apache.org/zE
> If you are an issue owner please let me know whether yours
> could be considered a blocker for 0.2
>
> Thanks,
> Roman.
>
Re: Giraph 0.2.0
Posted by Avery Ching <ac...@apache.org>.
Thanks for volunteering to do this Roman! It would be a great help for
Giraph users.
On 10/22/12 4:28 PM, Roman Shaposhnik wrote:
> Hi!
>
> As was discussed on GIRAPH-285 it seems to be right time
> to think about the release. It seems like Avery has agreed
> to be an official RM and I'm volunteering for an RM sidekick
> role ;-)
>
> My plan is pretty simple: I would like to start with JIRA triaging,
> figure out which ones folks consider to be blockers for 0.2 and
> then help Avery curate the release branch. I am also volunteering
> to test RCs with Hadoop 2.0.2-alpha (as part of the Bigtop 0.5.0)
> and also do some limited testing with Hadoop 1.1.0 and CDH3/4
>
> If this sounds like a good plan, here's a couple of areas where
> help would still be appreciated:
> * if you can test RCs with different versions of Hadoop -- please
> let me know. In fact, my most rigorous testing (to the extent
> it can be called that way) will be focused on Hadoop 2.0.2-alpha
>
> * JIRA triage -- we've got 17 issues tagged with 0.2
> http://s.apache.org/qH
> and 22 issues that are assigned to folks, but not to the
> 0.2 release:
> http://s.apache.org/zE
> If you are an issue owner please let me know whether yours
> could be considered a blocker for 0.2
>
> Thanks,
> Roman.