postgres-xc-general Mailing List for Postgres-XC

Hi Pavan,

 Sounds great.  I am interested to take a second look at XL.  Will try to
redo the comparison
in light of the performance improvements to XL.


Thanks
Sandeep




On Thu, Jan 21, 2016 at 3:39 AM, Pavan Deolasee <pav...@gm...>
wrote:

> Hi Sandeep,
>
> We're working to consolidate XC and XL efforts, but it may take some time.
> In the meanwhile, both XC and XL are being enhanced independently, often
> leveraging from each other's work.
>
> We've recently fixed several performance related issues in XL, especially
> for OLTP-like workload and I'm quite hopeful that the gap between XC and XL
> for such workloads must have drastically reduced, if not crossed already.
> For OLAP complex queries, XL should generally do better than XC.
>
> Do you mind testing the latest XL 9.5 code base in your environment and
> let us know if the performance is satisfactory? Our goal is to keep
> improving performance for both OLAP as well as OLTP workloads. So if there
> are gaps, those must be filled. The latest code is available here and I
> would suggest using XL9_5_STABLE branch for the tests.
> git://git.postgresql.org/git/postgres-xl.git
>
> Thanks,
> Pavan
>
> On Thu, Jan 21, 2016 at 6:47 AM, Sandeep Gupta <gup...@gm...>
> wrote:
>
>> Hi,
>>
>> In the past we  had used postgres-xc to successfully
>> scale our workload which mostly consisted of simple
>> filter-join-groupby expression.
>> All this required some careful data distribution strategies but on
>> average we were
>> able to extract good performance.
>>
>>  We tested our workloads with XL and the performance, at least for our
>> workload, was
>>  not at par. In this regard, I was wondering what is the future
>> direction of XC.
>>  If it is shelved, then shouldn't there be a way to match the
>> performance in XL given
>>  that XL has more generic framework compared to XC.
>>
>>
>>
>>  Best,
>>  Sandeep
>>
>>
>> ------------------------------------------------------------------------------
>> Site24x7 APM Insight: Get Deep Visibility into Application Performance
>> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
>> Monitor end-to-end web transactions and take corrective actions now
>> Troubleshoot faster and improve end-user experience. Signup Now!
>> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
>> _______________________________________________
>> Postgres-xc-general mailing list
>> Pos...@li...
>> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general
>>
>
>
>
> --
>  Pavan Deolasee                   http://www.2ndQuadrant.com/
>  PostgreSQL Development, 24x7 Support, Training & Services
>

Hi Sandeep,

We're working to consolidate XC and XL efforts, but it may take some time.
In the meanwhile, both XC and XL are being enhanced independently, often
leveraging from each other's work.

We've recently fixed several performance related issues in XL, especially
for OLTP-like workload and I'm quite hopeful that the gap between XC and XL
for such workloads must have drastically reduced, if not crossed already.
For OLAP complex queries, XL should generally do better than XC.

Do you mind testing the latest XL 9.5 code base in your environment and let
us know if the performance is satisfactory? Our goal is to keep improving
performance for both OLAP as well as OLTP workloads. So if there are gaps,
those must be filled. The latest code is available here and I would suggest
using XL9_5_STABLE branch for the tests.
git://git.postgresql.org/git/postgres-xl.git

Thanks,
Pavan

On Thu, Jan 21, 2016 at 6:47 AM, Sandeep Gupta <gup...@gm...>
wrote:

> Hi,
>
> In the past we  had used postgres-xc to successfully
> scale our workload which mostly consisted of simple
> filter-join-groupby expression.
> All this required some careful data distribution strategies but on
> average we were
> able to extract good performance.
>
>  We tested our workloads with XL and the performance, at least for our
> workload, was
>  not at par. In this regard, I was wondering what is the future
> direction of XC.
>  If it is shelved, then shouldn't there be a way to match the
> performance in XL given
>  that XL has more generic framework compared to XC.
>
>
>
>  Best,
>  Sandeep
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
> _______________________________________________
> Postgres-xc-general mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general
>



-- 
 Pavan Deolasee                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

Hi,

In the past we  had used postgres-xc to successfully
scale our workload which mostly consisted of simple
filter-join-groupby expression.
All this required some careful data distribution strategies but on
average we were
able to extract good performance.

 We tested our workloads with XL and the performance, at least for our
workload, was
 not at par. In this regard, I was wondering what is the future
direction of XC.
 If it is shelved, then shouldn't there be a way to match the
performance in XL given
 that XL has more generic framework compared to XC.



 Best,
 Sandeep