Re: GIN improvements part 1: additional information

On Fri, Jan 24, 2014 at 1:19 PM, Heikki Linnakangas <hlinnakangas(at)vmware(dot)com
> wrote:

> On 01/24/2014 10:53 AM, Alexander Korotkov wrote:
>
>> OK. What about previous fix in assert?
>>
>
> Ah right, fixed that too now.

Good, now my test-suite passed. Results are so.

Time of operations
event | period
-----------------------+-----------------
index_build | 00:01:45.709546
index_build_recovery | 00:00:09
index_update | 00:05:45.732214
index_update_recovery | 00:01:17
search_new | 00:24:02.191049
search_updated | 00:26:54.122852
(6 rows)

Index sizes
label | size
---------------+-----------
new | 387547136
after_updates | 610877440
(2 rows)

Before compression results were following.

Time of operations
event | period
-----------------------+-----------------
index_build | 00:01:51.116722
index_build_recovery | 00:00:08
index_update | 00:05:12.124758
index_update_recovery | 00:01:43
search_new | 00:23:44.281424
search_updated | 00:25:41.96251
(6 rows)

Index sizes
label | size
---------------+------------
new | 884514816
after_updates | 1595252736
(2 rows)

So, we can see little regression. However, I'm not sure if it's very
significant.

------
With best regards,
Alexander Korotkov.