Re: Research/Implementation of Nested Loop Join optimization

Tom Lane wrote:
> Gregory Stark <stark@enterprisedb.com> writes:
>> "Manoel Henrique" <mhenriquesgbd@gmail.com> writes:
>>> Yes, I'm relying on the assumption that backwards scan has the same cost as
>>> forward scan, why shouldn't it?
> 
> G...we expect that forward scans will result
> in the kernel doing read-ahead, ...
> A backwards scan will get no such overlapping and thus be up to 2X
> slower, unless the kernel is smart enough to do read-ahead for
> descending-order read requests.  Which seems not too probable. 

Linux's old adaptive readahead patches claimed to[1]:  It also have methods to detect some less common cases:  -
readingbackward"
 
Interestingly the author of that patch used postgres as the example
application that benefits from the patch (30%).

I'm not sure if the backward reading feature got kept
in the simplified on-demand readahead that seems to have
superseded the adaptive readahead stuff in 2.6.23[2].

[1] http://lwn.net/Articles/185469/
[2] http://kernelnewbies.org/Linux_2_6_23#head-102af265937262a7a21766ae58fddc1a29a5d8d7