On Fri, 15 Nov 2019 at 11:54, Amit Kapila wrote:
>
> On Wed, Nov 13, 2019 at 10:02 AM Masahiko Sawada
> wrote:
> >
> > I've done some tests while changing shared buffer size, delays and
> > number of workers. The overall results has the similar tendency as the
> > result shared by Dilip and looks
On Thu, Nov 14, 2019 at 5:02 PM Mahendra Singh wrote:
>
> On Mon, 11 Nov 2019 at 17:56, Amit Kapila wrote:
> >
> > On Mon, Nov 11, 2019 at 5:14 PM Dilip Kumar wrote:
> > >
> > > On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila
> > > wrote:
> > > >
> > > > ..
> > > > > I have tested the same with so
On Wed, Nov 13, 2019 at 10:02 AM Masahiko Sawada
wrote:
>
> I've done some tests while changing shared buffer size, delays and
> number of workers. The overall results has the similar tendency as the
> result shared by Dilip and looks reasonable to me.
>
Thanks, Sawada-san for repeating the tests
On Mon, 11 Nov 2019 at 17:56, Amit Kapila wrote:
>
> On Mon, Nov 11, 2019 at 5:14 PM Dilip Kumar wrote:
> >
> > On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila
wrote:
> > >
> > > ..
> > > > I have tested the same with some other workload(test file attached).
> > > > I can see the same behaviour with
On Tue, 12 Nov 2019 at 20:22, Masahiko Sawada
wrote:
>
> On Tue, 12 Nov 2019 at 19:08, Amit Kapila wrote:
> >
> > On Tue, Nov 12, 2019 at 3:03 PM Dilip Kumar wrote:
> > >
> > > On Tue, Nov 12, 2019 at 10:47 AM Dilip Kumar
> > > wrote:
> > > >
> > > > On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila
On Tue, 12 Nov 2019 at 19:08, Amit Kapila wrote:
>
> On Tue, Nov 12, 2019 at 3:03 PM Dilip Kumar wrote:
> >
> > On Tue, Nov 12, 2019 at 10:47 AM Dilip Kumar wrote:
> > >
> > > On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila
> > > wrote:
> > > >
> > > > On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar
On Tue, Nov 12, 2019 at 3:03 PM Dilip Kumar wrote:
>
> On Tue, Nov 12, 2019 at 10:47 AM Dilip Kumar wrote:
> >
> > On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila wrote:
> > >
> > > On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar
> > > wrote:
> > > >
> > > > On Mon, Nov 11, 2019 at 9:43 AM Dilip Kuma
On Tue, Nov 12, 2019 at 10:47 AM Dilip Kumar wrote:
>
> On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila wrote:
> >
> > On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar wrote:
> > >
> > > On Mon, Nov 11, 2019 at 9:43 AM Dilip Kumar wrote:
> > > >
> > > > On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila
> >
On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila wrote:
>
> On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar wrote:
> >
> > On Mon, Nov 11, 2019 at 9:43 AM Dilip Kumar wrote:
> > >
> > > On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila
> > > wrote:
> > > >
> > > >
> > > > Yeah, I think it is difficult to ge
On Mon, Nov 11, 2019 at 5:14 PM Dilip Kumar wrote:
>
> On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila wrote:
> >
> > ..
> > > I have tested the same with some other workload(test file attached).
> > > I can see the same behaviour with this workload as well that with the
> > > patch 4 the distributio
On Mon, Nov 11, 2019 at 4:23 PM Amit Kapila wrote:
>
> On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar wrote:
> >
> > On Mon, Nov 11, 2019 at 9:43 AM Dilip Kumar wrote:
> > >
> > > On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila
> > > wrote:
> > > >
> > > >
> > > > Yeah, I think it is difficult to ge
On Mon, Nov 11, 2019 at 12:59 PM Dilip Kumar wrote:
>
> On Mon, Nov 11, 2019 at 9:43 AM Dilip Kumar wrote:
> >
> > On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila wrote:
> > >
> > >
> > > Yeah, I think it is difficult to get the exact balance, but we can try
> > > to be as close as possible. We can
On Mon, Nov 11, 2019 at 9:43 AM Dilip Kumar wrote:
>
> On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila wrote:
> >
> > On Fri, Nov 8, 2019 at 9:39 AM Dilip Kumar wrote:
> > >
> > > I have done some experiments on this line. I have first produced a
> > > case where we can show the problem with the ex
On Fri, Nov 8, 2019 at 11:49 AM Amit Kapila wrote:
>
> On Fri, Nov 8, 2019 at 9:39 AM Dilip Kumar wrote:
> >
> > I have done some experiments on this line. I have first produced a
> > case where we can show the problem with the existing shared costing
> > patch (worker which is doing less I/O mi
On Fri, Nov 8, 2019 at 9:39 AM Dilip Kumar wrote:
>
> I have done some experiments on this line. I have first produced a
> case where we can show the problem with the existing shared costing
> patch (worker which is doing less I/O might pay the penalty on behalf
> of the worker who is doing more
On Fri, Nov 8, 2019 at 8:37 AM Amit Kapila wrote:
>> On Fri, Nov 8, 2019 at 8:18 AM Masahiko Sawada
> wrote:
> >
> > On Wed, 6 Nov 2019 at 15:45, Amit Kapila wrote:
> > >
> > > On Tue, Nov 5, 2019 at 11:28 AM Amit Kapila
> > > wrote:
> > > >
> > > > On Mon, Nov 4, 2019 at 11:42 PM Andres Freun
On Fri, Nov 8, 2019 at 8:18 AM Masahiko Sawada
wrote:
>
> On Wed, 6 Nov 2019 at 15:45, Amit Kapila wrote:
> >
> > On Tue, Nov 5, 2019 at 11:28 AM Amit Kapila wrote:
> > >
> > > On Mon, Nov 4, 2019 at 11:42 PM Andres Freund wrote:
> > > >
> > > >
> > > > > The two approaches to solve this proble
On Wed, 6 Nov 2019 at 15:45, Amit Kapila wrote:
>
> On Tue, Nov 5, 2019 at 11:28 AM Amit Kapila wrote:
> >
> > On Mon, Nov 4, 2019 at 11:42 PM Andres Freund wrote:
> > >
> > >
> > > > The two approaches to solve this problem being discussed in that
> > > > thread [1] are as follows:
> > > > (a)
On Wed, Nov 6, 2019 at 9:21 AM Stephen Frost wrote:
>
> Greetings,
>
> * Amit Kapila (amit.kapil...@gmail.com) wrote:
> > On Tue, Nov 5, 2019 at 1:42 AM Stephen Frost wrote:
> > > * Andres Freund (and...@anarazel.de) wrote:
> > > > That's quite doable independent of parallelism, as we don't have
On Tue, Nov 5, 2019 at 11:28 AM Amit Kapila wrote:
>
> On Mon, Nov 4, 2019 at 11:42 PM Andres Freund wrote:
> >
> >
> > > The two approaches to solve this problem being discussed in that
> > > thread [1] are as follows:
> > > (a) Allow the parallel workers and master backend to have a shared
> >
Greetings,
* Amit Kapila (amit.kapil...@gmail.com) wrote:
> On Tue, Nov 5, 2019 at 1:42 AM Stephen Frost wrote:
> > * Andres Freund (and...@anarazel.de) wrote:
> > > That's quite doable independent of parallelism, as we don't have tables
> > > or indexes spanning more than one tablespace. True,
On Wed, Nov 6, 2019 at 7:55 AM Andres Freund wrote:
>
> Hi,
>
> On 2019-11-06 07:53:09 +0530, Amit Kapila wrote:
> > As per feedback in this thread, it seems that for now, it is better,
> > if we can allow a parallel vacuum only when I/O throttling is not
> > enabled. We can later extend it based
Hi,
On 2019-11-06 07:53:09 +0530, Amit Kapila wrote:
> As per feedback in this thread, it seems that for now, it is better,
> if we can allow a parallel vacuum only when I/O throttling is not
> enabled. We can later extend it based on feedback from the field once
> the feature starts getting used
On Tue, Nov 5, 2019 at 1:42 AM Stephen Frost wrote:
> * Andres Freund (and...@anarazel.de) wrote:
>
> > That's quite doable independent of parallelism, as we don't have tables
> > or indexes spanning more than one tablespace. True, you could then make
> > the processing of an individual vacuum fa
On Tue, Nov 5, 2019 at 8:49 PM Andres Freund wrote:
>
> Hi,
>
> On November 5, 2019 7:16:41 AM PST, Dilip Kumar wrote:
> >On Tue, Nov 5, 2019 at 2:40 PM Amit Kapila
> >wrote:
> >>
> >> On Mon, Nov 4, 2019 at 11:58 PM Andres Freund
> >wrote:
> >> >
> >> > Hi,
> >> >
> >> > On 2019-11-04 12:59:02
Hi,
On November 5, 2019 7:16:41 AM PST, Dilip Kumar wrote:
>On Tue, Nov 5, 2019 at 2:40 PM Amit Kapila
>wrote:
>>
>> On Mon, Nov 4, 2019 at 11:58 PM Andres Freund
>wrote:
>> >
>> > Hi,
>> >
>> > On 2019-11-04 12:59:02 -0500, Jeff Janes wrote:
>> > > On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila
>
On Tue, Nov 5, 2019 at 2:40 PM Amit Kapila wrote:
>
> On Mon, Nov 4, 2019 at 11:58 PM Andres Freund wrote:
> >
> > Hi,
> >
> > On 2019-11-04 12:59:02 -0500, Jeff Janes wrote:
> > > On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila
> > > wrote:
> > >
> > > > For parallel vacuum [1], we were discussing
On Tue, Nov 5, 2019 at 1:12 AM Andres Freund wrote:
> On 2019-11-04 14:33:41 -0500, Stephen Frost wrote:
>
> > I've been wondering if the accounting system should consider the cost
> > per tablespace when there's multiple tablespaces involved, instead of
> > throttling the overall process without
On Mon, Nov 4, 2019 at 11:58 PM Andres Freund wrote:
>
> Hi,
>
> On 2019-11-04 12:59:02 -0500, Jeff Janes wrote:
> > On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila wrote:
> >
> > > For parallel vacuum [1], we were discussing what is the best way to
> > > divide the cost among parallel workers but we
On Mon, Nov 4, 2019 at 11:42 PM Andres Freund wrote:
>
>
> > The two approaches to solve this problem being discussed in that
> > thread [1] are as follows:
> > (a) Allow the parallel workers and master backend to have a shared
> > view of vacuum cost related parameters (mainly VacuumCostBalance)
Greetings,
* Andres Freund (and...@anarazel.de) wrote:
> On 2019-11-04 14:33:41 -0500, Stephen Frost wrote:
> > * Andres Freund (and...@anarazel.de) wrote:
> > > On 2019-11-04 14:06:19 -0500, Stephen Frost wrote:
> > > > With parallelization across indexes, you could have a situation where
> > > >
Hi,
On 2019-11-04 14:33:41 -0500, Stephen Frost wrote:
> * Andres Freund (and...@anarazel.de) wrote:
> > On 2019-11-04 14:06:19 -0500, Stephen Frost wrote:
> > > With parallelization across indexes, you could have a situation where
> > > the individual indexes are on different tablespaces with ind
Greetings,
* Andres Freund (and...@anarazel.de) wrote:
> On 2019-11-04 14:06:19 -0500, Stephen Frost wrote:
> > * Jeff Janes (jeff.ja...@gmail.com) wrote:
> > > On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila
> > > wrote:
> > > > For parallel vacuum [1], we were discussing what is the best way to
> >
Hi,
On 2019-11-04 14:06:19 -0500, Stephen Frost wrote:
> * Jeff Janes (jeff.ja...@gmail.com) wrote:
> > On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila wrote:
> > > For parallel vacuum [1], we were discussing what is the best way to
> > > divide the cost among parallel workers but we didn't get many i
Greetings,
* Jeff Janes (jeff.ja...@gmail.com) wrote:
> On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila wrote:
> > For parallel vacuum [1], we were discussing what is the best way to
> > divide the cost among parallel workers but we didn't get many inputs
> > apart from people who are very actively in
Hi,
On 2019-11-04 12:59:02 -0500, Jeff Janes wrote:
> On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila wrote:
>
> > For parallel vacuum [1], we were discussing what is the best way to
> > divide the cost among parallel workers but we didn't get many inputs
> > apart from people who are very actively in
Hi,
On 2019-11-04 12:24:35 +0530, Amit Kapila wrote:
> For parallel vacuum [1], we were discussing what is the best way to
> divide the cost among parallel workers but we didn't get many inputs
> apart from people who are very actively involved in patch development.
> I feel that we need some more
On Mon, Nov 4, 2019 at 1:54 AM Amit Kapila wrote:
> For parallel vacuum [1], we were discussing what is the best way to
> divide the cost among parallel workers but we didn't get many inputs
> apart from people who are very actively involved in patch development.
> I feel that we need some more i
On Mon, 4 Nov 2019 at 19:26, Amit Kapila wrote:
>
> On Mon, Nov 4, 2019 at 1:51 PM Masahiko Sawada wrote:
> >
> > On Mon, Nov 4, 2019 at 3:54 PM Amit Kapila wrote:
> > >
> > > I think approach-2 is better in throttling the system as it doesn't
> > > have the drawback of the first approach, but i
On Mon, Nov 4, 2019 at 1:03 PM Darafei "Komяpa" Praliaskouski
wrote:
>>
>>
>> This is somewhat similar to a memory usage problem with a
>> parallel query where each worker is allowed to use up to work_mem of
>> memory. We can say that the users using parallel operation can expect
>> more system r
On Mon, Nov 4, 2019 at 1:51 PM Masahiko Sawada wrote:
>
> On Mon, Nov 4, 2019 at 3:54 PM Amit Kapila wrote:
> >
> > I think approach-2 is better in throttling the system as it doesn't
> > have the drawback of the first approach, but it might be a bit tricky
> > to implement.
>
> I might be missin
On Mon, Nov 4, 2019 at 3:54 PM Amit Kapila wrote:
>
> I think approach-2 is better in throttling the system as it doesn't
> have the drawback of the first approach, but it might be a bit tricky
> to implement.
I might be missing something but I think that there could be the
drawback of the approa
>
>
> This is somewhat similar to a memory usage problem with a
> parallel query where each worker is allowed to use up to work_mem of
> memory. We can say that the users using parallel operation can expect
> more system resources to be used as they want to get the operation
> done faster, so we a
For parallel vacuum [1], we were discussing what is the best way to
divide the cost among parallel workers but we didn't get many inputs
apart from people who are very actively involved in patch development.
I feel that we need some more inputs before we finalize anything, so
starting a new thread.
44 matches
Mail list logo
