Doc patch for index access method function

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Doc patch for index access method function

Tatsuro Yamada
Hi,

Attached patch for fixing documents of "61.2. Index Access Method Functions" and
"61.6. Index Cost Estimation Functions".

I added a variable "double *indexPages" introduced by commit 5262f7a4f to the documents and
also added its explanation. Please read and revise it because I'm a non-native English speaker.

Regards,

Tatsuro Yamada
NTT Open Source Software Center


fix_index_access_method_doc.patch (2K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Doc patch for index access method function

Alexander Korotkov
Hi!

On Fri, Aug 10, 2018 at 6:24 AM Tatsuro Yamada
<[hidden email]> wrote:
>
> Attached patch for fixing documents of "61.2. Index Access Method Functions" and
> "61.6. Index Cost Estimation Functions".
>
> I added a variable "double *indexPages" introduced by commit 5262f7a4f to the documents and
> also added its explanation. Please read and revise it because I'm a non-native English speaker.

Good catch.  It was overseen in 5262f7a4fc44, where parallel index
scan was introduced.

"This is used to adjust the estimate for the cost of the disk access."

This sentence doesn't look correct for me.  Cost of the disk access is
estimated inside amcostestimate().  As I get, indexPages is used to
estimate how effective parallel scan would be, because different
workers pick different leaf pages.  I'm going to adjust this sentence
and commit this fix.

------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company

Reply | Threaded
Open this post in threaded view
|

Re: Doc patch for index access method function

Alexander Korotkov
On Fri, Aug 10, 2018 at 1:37 PM Alexander Korotkov
<[hidden email]> wrote:

> On Fri, Aug 10, 2018 at 6:24 AM Tatsuro Yamada
> <[hidden email]> wrote:
> >
> > Attached patch for fixing documents of "61.2. Index Access Method Functions" and
> > "61.6. Index Cost Estimation Functions".
> >
> > I added a variable "double *indexPages" introduced by commit 5262f7a4f to the documents and
> > also added its explanation. Please read and revise it because I'm a non-native English speaker.
>
> Good catch.  It was overseen in 5262f7a4fc44, where parallel index
> scan was introduced.
>
> "This is used to adjust the estimate for the cost of the disk access."
>
> This sentence doesn't look correct for me.  Cost of the disk access is
> estimated inside amcostestimate().  As I get, indexPages is used to
> estimate how effective parallel scan would be, because different
> workers pick different leaf pages.  I'm going to adjust this sentence
> and commit this fix.

Pushed.

------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company

Reply | Threaded
Open this post in threaded view
|

Re: Doc patch for index access method function

Tatsuro Yamada
Hi Alexander,

On 2018/08/10 20:27, Alexander Korotkov wrote:

> On Fri, Aug 10, 2018 at 1:37 PM Alexander Korotkov
> <[hidden email]> wrote:
>> On Fri, Aug 10, 2018 at 6:24 AM Tatsuro Yamada
>> <[hidden email]> wrote:
>>>
>>> Attached patch for fixing documents of "61.2. Index Access Method Functions" and
>>> "61.6. Index Cost Estimation Functions".
>>>
>>> I added a variable "double *indexPages" introduced by commit 5262f7a4f to the documents and
>>> also added its explanation. Please read and revise it because I'm a non-native English speaker.
>>
>> Good catch.  It was overseen in 5262f7a4fc44, where parallel index
>> scan was introduced.
>>
>> "This is used to adjust the estimate for the cost of the disk access."
>>
>> This sentence doesn't look correct for me.  Cost of the disk access is
>> estimated inside amcostestimate().  As I get, indexPages is used to
>> estimate how effective parallel scan would be, because different
>> workers pick different leaf pages.  I'm going to adjust this sentence
>> and commit this fix.
>
> Pushed.

I agree the sentence you corrected.
Thanks for taking your time!

Tatsuro Yamada
NTT Open Source Software Center



Previous Thread Next Thread