PerformanceEvaluation: filterScan

[Jean-Marc Spaggiari <je...@spaggiari.org>]

Any one tries this recently?

hbase org.apache.hadoop.hbase.PerformanceEvaluation --nomapred --rows=100
filterScan 3

I let it run for 30 minutes and killed it. 0 requets against the table, 649
requests against the META.

Seems to not be working anymore. Did not get a chance to look at the code
yet. Will look at the other options of PE against 1.0.0.

JM

Re: PerformanceEvaluation: filterScan

[ramkrishna vasudevan <ra...@gmail.com>]

I am seeing that if I run scanRange with multiThreads  (for eg : 25) it is
never ending. The reason may be that for every thread we try to form start
and stop row for every row in the input list.

So for measuring the latency for RangeScan10000Test it is better for us to
calculate the time per thread per scan operation and then do an average
ourselves.

Regards
Ram

On Tue, Feb 17, 2015 at 5:39 AM, Jerry He <je...@gmail.com> wrote:

> Hi,
>
> I was on 0.98 running PerformanceEvaluation.  randomRead or sequentialRead
> all show good read counts on the table regions.
>
> filterScan shows exactly the same as you had.
>
> Looking at the code.  It could be as expected.  The filter has no matching
> returns, so the region server is in nextInternal() loop.  And it seems that
> we only update the read request count for return rows?
>
> Jerry
>
> On Mon, Feb 16, 2015 at 6:43 AM, Jean-Marc Spaggiari <
> jean-marc@spaggiari.org> wrote:
>
> > Any one tries this recently?
> >
> > hbase org.apache.hadoop.hbase.PerformanceEvaluation --nomapred --rows=100
> > filterScan 3
> >
> > I let it run for 30 minutes and killed it. 0 requets against the table,
> 649
> > requests against the META.
> >
> > Seems to not be working anymore. Did not get a chance to look at the code
> > yet. Will look at the other options of PE against 1.0.0.
> >
> > JM
> >
>

Re: PerformanceEvaluation: filterScan

[Jerry He <je...@gmail.com>]

Hi,

I was on 0.98 running PerformanceEvaluation.  randomRead or sequentialRead
all show good read counts on the table regions.

filterScan shows exactly the same as you had.

Looking at the code.  It could be as expected.  The filter has no matching
returns, so the region server is in nextInternal() loop.  And it seems that
we only update the read request count for return rows?

Jerry

On Mon, Feb 16, 2015 at 6:43 AM, Jean-Marc Spaggiari <
jean-marc@spaggiari.org> wrote:

> Any one tries this recently?
>
> hbase org.apache.hadoop.hbase.PerformanceEvaluation --nomapred --rows=100
> filterScan 3
>
> I let it run for 30 minutes and killed it. 0 requets against the table, 649
> requests against the META.
>
> Seems to not be working anymore. Did not get a chance to look at the code
> yet. Will look at the other options of PE against 1.0.0.
>
> JM
>