git.net

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Releasing and publishing a Calcite Benchmark and Performance Profiler


Thank you.

Let's have a videoconference in couple of weeks to discuss a progress to
date, and what else is left to do to have a valuable benchmark and a
profiling approach.

I will send an invite to this list with couple of suggested times.

Thank you,
Edmon

On Fri, May 11, 2018 at 9:06 AM, Ruhollah Farchtchi <ruhollah@xxxxxxxxxxxx>
wrote:

> Edmon and Michael, I would be able to have some of our developers at
> Zoomdata help with this effort as well. Also, I'm happy to co-author any
> papers on the results. We've already been exploring using Zipkin for some
> of our other initiatives so this effort lines up quite nicely and the
> timeline presented works for us too.
>
> *Ruhollah Farchtchi*
> CTO Zoomdata
> ruhollah@xxxxxxxxxxxx
> 703-944-9791
> www.zoomdata.com
>
> On Thu, May 10, 2018 at 9:34 AM, Michael Mior <mmior@xxxxxxxxxxxx> wrote:
>
> > I should be able to have some time over the next couple months to pitch
> in.
> > Instrumenting Calcite would be a useful endeavour on its own. Perhaps
> with
> > something like Zipkin? I think it would be interesting to talk about what
> > we can do with adapters as well.
> >
> > --
> > Michael Mior
> > mmior@xxxxxxxxxxxx
> >
> >
> > Le jeu. 10 mai 2018 à 05:08, Edmon Begoli <ebegoli@xxxxxxxxxx> a écrit :
> >
> > > Colleagues,
> > >
> > > My R&D team and I are working towards ver. 1.0 of the Calcite
> Benchmark.
> > >
> > > Does anyone have time and energy to contribute in whatever capacity,
> > > perhaps make benchmark more ambitious and maybe have a paper with us on
> > it
> > > targeting BigData 2018 with a due date of 8/8/2018?
> > >
> > > We have 4-5 pages of material written, but mostly on setup and for
> > > benchmarking Postgres through Calcite via TPC-DS.
> > >
> > > Valuable parts for contribution would be to:
> > >
> > > a) instrument and measure Calcite overhead distribution,
> > > b) optimizer perfrormance
> > > c) mix in additional targets - Cassandra, Mongo or even Flink
> > >
> > > whatever else ...
> > >
> > > *Please let me know. *
> > >
> > > Thank you,
> > > Edmon
> > >
> >
>