You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@celeborn.apache.org by Keyong Zhou <zh...@apache.org> on 2022/12/03 09:44:30 UTC
[DISCUSS] Prepare first release
Hi celeborn (-incubating) community:
One and a half months has passed since celeborn entered Apache Incubator.
Although it's new to Apache, it's already widely deployed on production
environment. I think it's time to prepare for our first release.
After discussing with several developers, we have a preliminary agreement
that we can (almost) check out the branch-0.2.0-rc1 release branch from
main. If you have any questions or ideas, feel free to discuss on this
thread.
Thanks!
Keyong Zhou
Re: [DISCUSS] Prepare first release
Posted by Ethan Feng <et...@gmail.com>.
Roger that. The check will be done.
Ethan.
在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
> Hi celeborn (-incubating) community:
>
> Currently we are preparing for the first release (branch-0.2). To ensure
> code quality, I would like to test for core-path correctness and stability,
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> <na...@gmail.com> help test graceful shutdown and rolling upgrade? And
> could Ethan Feng <et...@gmail.com> help test load-aware slots
> allocation?
>
> And we would be rather happy if anyone can help test for other modules
> (k8s, HA, etc.).
>
> Thanks,
> Keyong
Re: [DISCUSS] Prepare first release
Posted by Keyong Zhou <zh...@apache.org>.
Thanks Angerszhuuuu <an...@gmail.com> !
I think we can start preparing the binaries Ethan Feng
<et...@gmail.com>
Thanks,
Keyong Zhou
angers zhu <an...@gmail.com> 于2023年1月3日周二 11:18写道：
> Hi all,
>
> I have checked the quick graceful shutdown and rolling restart worker. Fix
> some minor problem and it's ok.
>
> Best Regards
> Angerszhuuuu
>
> Keyong Zhou <zh...@apache.org> 于2022年12月21日周三 11:39写道：
>
> > Thank you guys for the valuable tests!
> >
> > angers zhu <an...@gmail.com> 于2022年12月21日周三 10:33写道：
> >
> > > Hi all,
> > >
> > > I have tested quota management and make some minor improvements. It can
> > > work well.
> > > later I will test worker quick restart.
> > >
> > > apache <et...@apache.org> 于2022年12月21日周三 10:00写道：
> > >
> > >> Hi community:
> > >> I’ve tested branch-0.2 about load-aware slots allocation modules with
> a
> > >> fusion cluster of HDDs, SSDs, and ESSDs. As expected, workers with
> > faster
> > >> disk drives have more slots on them.
> > >>
> > >> The way I tested is as follows:
> > >> I ran 1T TPC-DS while 1TB Terasort was running repeatedly and tested
> > >> different parameter combinations on three node groups. The group core
> > has
> > >> 10x NVME SSD drives. Group task 1 has 4x ESSD drives. Group task 2 has
> > 12x
> > >> SATA HDD drivers.
> > >>
> > >> Here are some snapshots:
> > >>
> > >> Graph 1: Only NVME SSD and HDD workers are involved and
> > >> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 2.
> > >>
> > >>
> > >> Graph 2: All nodes are involved and
> > >> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 3.
> > >>
> > >>
> > >> Thanks,
> > >> Ethan Feng
> > >> 在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
> > >>
> > >> Hi celeborn (-incubating) community:
> > >>
> > >> Currently we are preparing for the first release (branch-0.2). To
> ensure
> > >> code quality, I would like to test for core-path correctness and
> > >> stability,
> > >> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> > >> <na...@gmail.com> help test graceful shutdown and rolling
> upgrade?
> > >> And
> > >> could Ethan Feng <et...@gmail.com> help test load-aware
> > >> slots
> > >> allocation?
> > >>
> > >> And we would be rather happy if anyone can help test for other modules
> > >> (k8s, HA, etc.).
> > >>
> > >> Thanks,
> > >> Keyong
> > >>
> > >>
> >
>
Re: [DISCUSS] Prepare first release
Posted by angers zhu <an...@gmail.com>.
Hi all,
I have checked the quick graceful shutdown and rolling restart worker. Fix
some minor problem and it's ok.
Best Regards
Angerszhuuuu
Keyong Zhou <zh...@apache.org> 于2022年12月21日周三 11:39写道：
> Thank you guys for the valuable tests!
>
> angers zhu <an...@gmail.com> 于2022年12月21日周三 10:33写道：
>
> > Hi all,
> >
> > I have tested quota management and make some minor improvements. It can
> > work well.
> > later I will test worker quick restart.
> >
> > apache <et...@apache.org> 于2022年12月21日周三 10:00写道：
> >
> >> Hi community:
> >> I’ve tested branch-0.2 about load-aware slots allocation modules with a
> >> fusion cluster of HDDs, SSDs, and ESSDs. As expected, workers with
> faster
> >> disk drives have more slots on them.
> >>
> >> The way I tested is as follows:
> >> I ran 1T TPC-DS while 1TB Terasort was running repeatedly and tested
> >> different parameter combinations on three node groups. The group core
> has
> >> 10x NVME SSD drives. Group task 1 has 4x ESSD drives. Group task 2 has
> 12x
> >> SATA HDD drivers.
> >>
> >> Here are some snapshots:
> >>
> >> Graph 1: Only NVME SSD and HDD workers are involved and
> >> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 2.
> >>
> >>
> >> Graph 2: All nodes are involved and
> >> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 3.
> >>
> >>
> >> Thanks,
> >> Ethan Feng
> >> 在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
> >>
> >> Hi celeborn (-incubating) community:
> >>
> >> Currently we are preparing for the first release (branch-0.2). To ensure
> >> code quality, I would like to test for core-path correctness and
> >> stability,
> >> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> >> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> >> And
> >> could Ethan Feng <et...@gmail.com> help test load-aware
> >> slots
> >> allocation?
> >>
> >> And we would be rather happy if anyone can help test for other modules
> >> (k8s, HA, etc.).
> >>
> >> Thanks,
> >> Keyong
> >>
> >>
>
Re: [DISCUSS] Prepare first release
Posted by Keyong Zhou <zh...@apache.org>.
Thank you guys for the valuable tests!
angers zhu <an...@gmail.com> 于2022年12月21日周三 10:33写道：
> Hi all,
>
> I have tested quota management and make some minor improvements. It can
> work well.
> later I will test worker quick restart.
>
> apache <et...@apache.org> 于2022年12月21日周三 10:00写道：
>
>> Hi community:
>> I’ve tested branch-0.2 about load-aware slots allocation modules with a
>> fusion cluster of HDDs, SSDs, and ESSDs. As expected, workers with faster
>> disk drives have more slots on them.
>>
>> The way I tested is as follows:
>> I ran 1T TPC-DS while 1TB Terasort was running repeatedly and tested
>> different parameter combinations on three node groups. The group core has
>> 10x NVME SSD drives. Group task 1 has 4x ESSD drives. Group task 2 has 12x
>> SATA HDD drivers.
>>
>> Here are some snapshots:
>>
>> Graph 1: Only NVME SSD and HDD workers are involved and
>> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 2.
>>
>>
>> Graph 2: All nodes are involved and
>> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 3.
>>
>>
>> Thanks,
>> Ethan Feng
>> 在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
>>
>> Hi celeborn (-incubating) community:
>>
>> Currently we are preparing for the first release (branch-0.2). To ensure
>> code quality, I would like to test for core-path correctness and
>> stability,
>> could Angerszhuuuu <an...@gmail.com> and nafiyaix
>> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
>> And
>> could Ethan Feng <et...@gmail.com> help test load-aware
>> slots
>> allocation?
>>
>> And we would be rather happy if anyone can help test for other modules
>> (k8s, HA, etc.).
>>
>> Thanks,
>> Keyong
>>
>>
Re: [DISCUSS] Prepare first release
Posted by angers zhu <an...@gmail.com>.
Hi all,
I have tested quota management and make some minor improvements. It can
work well.
later I will test worker quick restart.
apache <et...@apache.org> 于2022年12月21日周三 10:00写道：
> Hi community:
> I’ve tested branch-0.2 about load-aware slots allocation modules with a
> fusion cluster of HDDs, SSDs, and ESSDs. As expected, workers with faster
> disk drives have more slots on them.
>
> The way I tested is as follows:
> I ran 1T TPC-DS while 1TB Terasort was running repeatedly and tested
> different parameter combinations on three node groups. The group core has
> 10x NVME SSD drives. Group task 1 has 4x ESSD drives. Group task 2 has 12x
> SATA HDD drivers.
>
> Here are some snapshots:
>
> Graph 1: Only NVME SSD and HDD workers are involved and
> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 2.
>
>
> Graph 2: All nodes are involved and
> "celeborn.slots.assign.loadAware.numDiskGroups" is set to 3.
>
>
> Thanks,
> Ethan Feng
> 在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
>
> Hi celeborn (-incubating) community:
>
> Currently we are preparing for the first release (branch-0.2). To ensure
> code quality, I would like to test for core-path correctness and stability,
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> And
> could Ethan Feng <et...@gmail.com> help test load-aware slots
> allocation?
>
> And we would be rather happy if anyone can help test for other modules
> (k8s, HA, etc.).
>
> Thanks,
> Keyong
>
>
Re: [DISCUSS] Prepare first release
Posted by apache <et...@apache.org>.
Hi community:
I’ve tested branch-0.2 about load-aware slots allocation modules with a fusion cluster of HDDs, SSDs, and ESSDs. As expected, workers with faster disk drives have more slots on them.
The way I tested is as follows:
I ran 1T TPC-DS while 1TB Terasort was running repeatedly and tested different parameter combinations on three node groups. The group core has 10x NVME SSD drives. Group task 1 has 4x ESSD drives. Group task 2 has 12x SATA HDD drivers.
Here are some snapshots:
Graph 1: Only NVME SSD and HDD workers are involved and "celeborn.slots.assign.loadAware.numDiskGroups" is set to 2.
Graph 2: All nodes are involved and "celeborn.slots.assign.loadAware.numDiskGroups" is set to 3.
Thanks,
Ethan Feng
在 2022年12月14日 +0800 19:41，Keyong Zhou <zh...@apache.org>，写道：
> Hi celeborn (-incubating) community:
>
> Currently we are preparing for the first release (branch-0.2). To ensure
> code quality, I would like to test for core-path correctness and stability,
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> <na...@gmail.com> help test graceful shutdown and rolling upgrade? And
> could Ethan Feng <et...@gmail.com> help test load-aware slots
> allocation?
>
> And we would be rather happy if anyone can help test for other modules
> (k8s, HA, etc.).
>
> Thanks,
> Keyong
Re: [DISCUSS] Prepare first release
Posted by keyong zhou <wa...@gmail.com>.
Hi Cheng,
Thanks for your test on k8s! I agree that we should fix the
configurations before 0.2.0.
Thanks,
Keyong Zhou
Cheng Pan <pa...@gmail.com> 于2022年12月20日周二 15:48写道：
> Hi community,
>
> I test branch-0.2 on K8s w/ CELEBORN-151 and some internal modification by
> running some TPC-DS quires, works as expected.
>
> I noticed that there are some configurations still use `rss.` prefix
> instead of `celeborn.`, we should fix them before releasing 0.2.0.
>
> Thanks,
> Cheng Pan
>
>
> On Dec 19, 2022 at 18:25:44, keyong zhou <wa...@gmail.com> wrote:
>
> > Hi Celeborn (Incubating) community,
> >
> > I have tested the core-path and fixed on bug[1]. After fixing this, my
> test
> > seems OK.
> >
> > The way I tested is as follows:
> > I run 1T TPCDS with Celeborn on, then I randomly kill worker during the
> > test, carefully not to trigger datalost. In the end I check whether the
> > results are correct.
> >
> > [1] https://github.com/apache/incubator-celeborn/pull/1101
> >
> > Thanks,
> > Keyong Zhou
> >
> > Keyong Zhou <zh...@apache.org> 于2022年12月14日周三 19:41写道：
> >
> > Hi celeborn (-incubating) community:
> >
> >
> > Currently we are preparing for the first release (branch-0.2). To ensure
> >
> > code quality, I would like to test for core-path correctness and
> stability,
> >
> > could Angerszhuuuu <an...@gmail.com> and nafiyaix
> >
> > <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> >
> > And could Ethan Feng <et...@gmail.com> help test load-aware
> >
> > slots allocation?
> >
> >
> > And we would be rather happy if anyone can help test for other modules
> >
> > (k8s, HA, etc.).
> >
> >
> > Thanks,
> >
> > Keyong
> >
> >
> >
>
Re: [DISCUSS] Prepare first release
Posted by Cheng Pan <pa...@gmail.com>.
Hi community,
I test branch-0.2 on K8s w/ CELEBORN-151 and some internal modification by
running some TPC-DS quires, works as expected.
I noticed that there are some configurations still use `rss.` prefix
instead of `celeborn.`, we should fix them before releasing 0.2.0.
Thanks,
Cheng Pan
On Dec 19, 2022 at 18:25:44, keyong zhou <wa...@gmail.com> wrote:
> Hi Celeborn (Incubating) community,
>
> I have tested the core-path and fixed on bug[1]. After fixing this, my test
> seems OK.
>
> The way I tested is as follows:
> I run 1T TPCDS with Celeborn on, then I randomly kill worker during the
> test, carefully not to trigger datalost. In the end I check whether the
> results are correct.
>
> [1] https://github.com/apache/incubator-celeborn/pull/1101
>
> Thanks,
> Keyong Zhou
>
> Keyong Zhou <zh...@apache.org> 于2022年12月14日周三 19:41写道：
>
> Hi celeborn (-incubating) community:
>
>
> Currently we are preparing for the first release (branch-0.2). To ensure
>
> code quality, I would like to test for core-path correctness and stability,
>
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
>
> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
>
> And could Ethan Feng <et...@gmail.com> help test load-aware
>
> slots allocation?
>
>
> And we would be rather happy if anyone can help test for other modules
>
> (k8s, HA, etc.).
>
>
> Thanks,
>
> Keyong
>
>
>
Re: [DISCUSS] Prepare first release
Posted by keyong zhou <wa...@gmail.com>.
Hi Celeborn (Incubating) community,
I have tested the core-path and fixed on bug[1]. After fixing this, my test
seems OK.
The way I tested is as follows:
I run 1T TPCDS with Celeborn on, then I randomly kill worker during the
test, carefully not to trigger datalost. In the end I check whether the
results are correct.
[1] https://github.com/apache/incubator-celeborn/pull/1101
Thanks,
Keyong Zhou
Keyong Zhou <zh...@apache.org> 于2022年12月14日周三 19:41写道：
> Hi celeborn (-incubating) community:
>
> Currently we are preparing for the first release (branch-0.2). To ensure
> code quality, I would like to test for core-path correctness and stability,
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> And could Ethan Feng <et...@gmail.com> help test load-aware
> slots allocation?
>
> And we would be rather happy if anyone can help test for other modules
> (k8s, HA, etc.).
>
> Thanks,
> Keyong
>
Re: [DISCUSS] Prepare first release
Posted by Keyong Zhou <zh...@apache.org>.
Thank you guys.
nafiyaix97 <na...@gmail.com> 于2022年12月15日周四 11:37写道：
> No problem, will check it with Angers.
>
> On Wed, Dec 14, 2022 at 7:41 PM Keyong Zhou <zh...@apache.org> wrote:
>
> > Hi celeborn (-incubating) community:
> >
> > Currently we are preparing for the first release (branch-0.2). To ensure
> > code quality, I would like to test for core-path correctness and
> stability,
> > could Angerszhuuuu <an...@gmail.com> and nafiyaix
> > <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> > And
> > could Ethan Feng <et...@gmail.com> help test load-aware
> slots
> > allocation?
> >
> > And we would be rather happy if anyone can help test for other modules
> > (k8s, HA, etc.).
> >
> > Thanks,
> > Keyong
> >
>
Re: [DISCUSS] Prepare first release
Posted by nafiyaix97 <na...@gmail.com>.
No problem, will check it with Angers.
On Wed, Dec 14, 2022 at 7:41 PM Keyong Zhou <zh...@apache.org> wrote:
> Hi celeborn (-incubating) community:
>
> Currently we are preparing for the first release (branch-0.2). To ensure
> code quality, I would like to test for core-path correctness and stability,
> could Angerszhuuuu <an...@gmail.com> and nafiyaix
> <na...@gmail.com> help test graceful shutdown and rolling upgrade?
> And
> could Ethan Feng <et...@gmail.com> help test load-aware slots
> allocation?
>
> And we would be rather happy if anyone can help test for other modules
> (k8s, HA, etc.).
>
> Thanks,
> Keyong
>
Re: [DISCUSS] Prepare first release
Posted by Keyong Zhou <zh...@apache.org>.
Hi celeborn (-incubating) community:
Currently we are preparing for the first release (branch-0.2). To ensure
code quality, I would like to test for core-path correctness and stability,
could Angerszhuuuu <an...@gmail.com> and nafiyaix
<na...@gmail.com> help test graceful shutdown and rolling upgrade? And
could Ethan Feng <et...@gmail.com> help test load-aware slots
allocation?
And we would be rather happy if anyone can help test for other modules
(k8s, HA, etc.).
Thanks,
Keyong
Re: [DISCUSS] Prepare first release
Posted by wangsheng <22...@qq.com.INVALID>.
+1, looking forward to seeing 0.2.0 celeborn.
Best,
Binjie Yang
> 2022年12月3日 下午6:58，Cheng Pan <pa...@gmail.com> 写道：
>
> +1 for release, the branch name should be branch-0.2, keep consistent w/
> existed branch-0.1
>
> Thanks,
> Cheng Pan
>
>
> On Dec 3, 2022 at 17:44:30, Keyong Zhou <zh...@apache.org> wrote:
>
>> Hi celeborn (-incubating) community:
>>
>> One and a half months has passed since celeborn entered Apache Incubator.
>> Although it's new to Apache, it's already widely deployed on production
>> environment. I think it's time to prepare for our first release.
>>
>> After discussing with several developers, we have a preliminary agreement
>> that we can (almost) check out the branch-0.2.0-rc1 release branch from
>> main. If you have any questions or ideas, feel free to discuss on this
>> thread.
>>
>> Thanks!
>> Keyong Zhou
>>
Re: [DISCUSS] Prepare first release
Posted by Keyong Zhou <zh...@apache.org>.
Sure, thanks for the correction.
Thanks,
Keyong Zhou
Cheng Pan <pa...@gmail.com> 于2022年12月3日周六 18:58写道：
> +1 for release, the branch name should be branch-0.2, keep consistent w/
> existed branch-0.1
>
> Thanks,
> Cheng Pan
>
>
> On Dec 3, 2022 at 17:44:30, Keyong Zhou <zh...@apache.org> wrote:
>
> > Hi celeborn (-incubating) community:
> >
> > One and a half months has passed since celeborn entered Apache Incubator.
> > Although it's new to Apache, it's already widely deployed on production
> > environment. I think it's time to prepare for our first release.
> >
> > After discussing with several developers, we have a preliminary agreement
> > that we can (almost) check out the branch-0.2.0-rc1 release branch from
> > main. If you have any questions or ideas, feel free to discuss on this
> > thread.
> >
> > Thanks!
> > Keyong Zhou
> >
>
Re: [DISCUSS] Prepare first release
Posted by Cheng Pan <pa...@gmail.com>.
+1 for release, the branch name should be branch-0.2, keep consistent w/
existed branch-0.1
Thanks,
Cheng Pan
On Dec 3, 2022 at 17:44:30, Keyong Zhou <zh...@apache.org> wrote:
> Hi celeborn (-incubating) community:
>
> One and a half months has passed since celeborn entered Apache Incubator.
> Although it's new to Apache, it's already widely deployed on production
> environment. I think it's time to prepare for our first release.
>
> After discussing with several developers, we have a preliminary agreement
> that we can (almost) check out the branch-0.2.0-rc1 release branch from
> main. If you have any questions or ideas, feel free to discuss on this
> thread.
>
> Thanks!
> Keyong Zhou
>