Hi,

Is anybody benchmarking the performance of server-side aggregations through
co-processors in HBase? I am interested to know if HBase could potentially
be used to calculate real-time SQL-like aggregations at a good level of
performance (q < 200ms on high-load, big dataset scenario). Just curious to
know before I implement my own benchmarks.

Pere.

For our solution we are doing some aggregation on the server via
coprocessors. In general, for each row there are 8 columns: 7 columns
that contain numbers (for summation) and 1 column that contains a
hyperloglog counter (about 700bytes). Functionally, this solution
works well and ought to scale with the number of region servers.
However, the individual request performance leaves a little to be
desired. What we've seen is that to scan 40000 rows (aggregated into
3000 rows) takes about 4 seconds.

Our code is in it's early stages (unoptimized) so we hope to see some
significant performance improvements when we run our coprocessor under
a profiler. Our benchmarks were on underpowered machines (only 2gb
RAM) as well.

Hope this helps!

--Tom

On Thu, May 3, 2012 at 6:08 AM, Pere Ferrera <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Is anybody benchmarking the performance of server-side aggregations through
> co-processors in HBase? I am interested to know if HBase could potentially
> be used to calculate real-time SQL-like aggregations at a good level of
> performance (q < 200ms on high-load, big dataset scenario). Just curious to
> know before I implement my own benchmarks.
>
> Pere.

We're seen reasonable performance, with the caveat that you need to
parallelize the scan doing the aggregation. In our benchmarking, we have
the client scan each region in parallel and have a coprocessor aggregate
the row count and return a single row back (with the client then
totaling the counts it gets back). Here are the numbers we've seen when
aggregating 1 million rows, this with a slightly older hbase version
(~0.92):

Schema: 50col x 50bytes with compressible data
Regions     RowCount         RowCount with single binary filter
             Time (sec)       Time (sec)
   1          11.3             19.0
   4           3.5              5.6
  16           1.8              2.6
  32           1.2              1.8

Schema: 1col x 2500bytes with compressible data
Regions     RowCount         RowCount with single binary filter
             Time (sec)       Time (sec)
   1           7.0              7.0
   4           1.2              1.2
  16           0.7              0.7
  32           0.3              0.3

This is run on a four machine cluster with each machine having 4G Heap
and with the servers warmed-up (cached data).

Hope this helps.

     James
On 05/03/2012 08:01 AM, Tom Brown wrote:
> For our solution we are doing some aggregation on the server via
> coprocessors. In general, for each row there are 8 columns: 7 columns
> that contain numbers (for summation) and 1 column that contains a
> hyperloglog counter (about 700bytes). Functionally, this solution
> works well and ought to scale with the number of region servers.
> However, the individual request performance leaves a little to be
> desired. What we've seen is that to scan 40000 rows (aggregated into
> 3000 rows) takes about 4 seconds.
>
> Our code is in it's early stages (unoptimized) so we hope to see some
> significant performance improvements when we run our coprocessor under
> a profiler. Our benchmarks were on underpowered machines (only 2gb
> RAM) as well.
>
> Hope this helps!
>
> --Tom
>
> On Thu, May 3, 2012 at 6:08 AM, Pere Ferrera<[EMAIL PROTECTED]>  wrote:
>> Hi,
>>
>> Is anybody benchmarking the performance of server-side aggregations through
>> co-processors in HBase? I am interested to know if HBase could potentially
>> be used to calculate real-time SQL-like aggregations at a good level of
>> performance (q<  200ms on high-load, big dataset scenario). Just curious to
>> know before I implement my own benchmarks.
>>
>> Pere.

I did some experiments which compares scan, coprocessor and mapreduce
approach, in an ec2 environment.
You may find it interesting:
http://hbase-coprocessor-experiments.blogspot.com/2011/05/extending.html

Thanks,
Himanshu

On Thu, May 3, 2012 at 11:02 AM, James Taylor <[EMAIL PROTECTED]> wrote:
> We're seen reasonable performance, with the caveat that you need to
> parallelize the scan doing the aggregation. In our benchmarking, we have the
> client scan each region in parallel and have a coprocessor aggregate the row
> count and return a single row back (with the client then totaling the counts
> it gets back). Here are the numbers we've seen when aggregating 1 million
> rows, this with a slightly older hbase version (~0.92):
>
> Schema: 50col x 50bytes with compressible data
> Regions     RowCount         RowCount with single binary filter
>            Time (sec)       Time (sec)
>  1          11.3             19.0
>  4           3.5              5.6
>  16           1.8              2.6
>  32           1.2              1.8
>
> Schema: 1col x 2500bytes with compressible data
> Regions     RowCount         RowCount with single binary filter
>            Time (sec)       Time (sec)
>  1           7.0              7.0
>  4           1.2              1.2
>  16           0.7              0.7
>  32           0.3              0.3
>
> This is run on a four machine cluster with each machine having 4G Heap and
> with the servers warmed-up (cached data).
>
> Hope this helps.
>
>    James
>
>
>
> On 05/03/2012 08:01 AM, Tom Brown wrote:
>>
>> For our solution we are doing some aggregation on the server via
>> coprocessors. In general, for each row there are 8 columns: 7 columns
>> that contain numbers (for summation) and 1 column that contains a
>> hyperloglog counter (about 700bytes). Functionally, this solution
>> works well and ought to scale with the number of region servers.
>> However, the individual request performance leaves a little to be
>> desired. What we've seen is that to scan 40000 rows (aggregated into
>> 3000 rows) takes about 4 seconds.
>>
>> Our code is in it's early stages (unoptimized) so we hope to see some
>> significant performance improvements when we run our coprocessor under
>> a profiler. Our benchmarks were on underpowered machines (only 2gb
>> RAM) as well.
>>
>> Hope this helps!
>>
>> --Tom
>>
>> On Thu, May 3, 2012 at 6:08 AM, Pere Ferrera<[EMAIL PROTECTED]>
>>  wrote:
>>>
>>> Hi,
>>>
>>> Is anybody benchmarking the performance of server-side aggregations
>>> through
>>> co-processors in HBase? I am interested to know if HBase could
>>> potentially
>>> be used to calculate real-time SQL-like aggregations at a good level of
>>> performance (q<  200ms on high-load, big dataset scenario). Just curious
>>> to
>>> know before I implement my own benchmarks.
>>>
>>> Pere.
>
>