Re: Compression of full-page-writes - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Compression of full-page-writes
Date
Msg-id CAB7nPqTp5QAGtNR5rXFhsHMjQanrQVKs4eLged+9w46qUENNrw@mail.gmail.com
Whole thread Raw
In response to Re: Compression of full-page-writes  (Rahila Syed <rahilasyed90@gmail.com>)
Responses Re: Compression of full-page-writes
List pgsql-hackers
On Tue, Jun 10, 2014 at 11:49 PM, Rahila Syed <rahilasyed90@gmail.com> wrote:
> Hello ,
>
>
> In order to facilitate changing of compression algorithms  and to be able to
> recover using WAL records compressed with different compression algorithms,
> information about compression algorithm can be stored in WAL record.
>
> XLOG record header has 2 to 4 padding bytes in order to align the WAL
> record. This space can be used for  a new flag in order to store information
> about the compression algorithm used. Like the xl_info field of XlogRecord
> struct,  8 bits flag  can be constructed with the lower 4 bits of the flag
> used to indicate which backup block is compressed out of 0,1,2,3. Higher
> four bits can be used to indicate state of compression i.e
> off,lz4,snappy,pglz.
>
> The flag can be extended to incorporate more compression algorithms added in
> future if any.
>
> What is your opinion on this?
-1 for any additional bytes in WAL record to control such things,
having one single compression that we know performs well and relying
on it makes the life of user and developer easier.
-- 
Michael



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: "RETURNING PRIMARY KEY" syntax extension
Next
From: Tom Lane
Date:
Subject: Re: branch for 9.5?