Beefy Boxes and Bandwidth Generously Provided by pair Networks
Pathologically Eclectic Rubbish Lister

Re^3: Processing ~1 Trillion records

by clueless newbie (Chaplain)
on Oct 25, 2012 at 12:51 UTC ( #1000857=note: print w/replies, xml ) Need Help??

in reply to Re^2: Processing ~1 Trillion records
in thread Processing ~1 Trillion records

To quote Tim Bunce

- while(@row = $sth->fetchrow_array) { } " one column: 51,155 fetches per second " 20 columns: 24,032 fetches per second - while($row = $sth->fetchrow_arrayref) { } " one column: 58,653 fetches per second - approximately 12% faster " 20 columns: 49,390 fetches per second - approximately 51% faster - while($row = shift(@$rowcache) || shift(@{$rowcache=$sth->fetchall_a +rrayref(undef, $max_rows)})) { } " one column: 348,140 fetches per second - by far the fastest! " 20 columns: 42,128 fetches per second - now slower than fetchrow_ +arrayref! " Why? Balance time saved making fewer calls with time spent managi +ng more memory " Do your own benchmarks to find what works best for your situation +s

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://1000857]
[shmem]: ...for espresso
[LanX]: due coretti, per favore! ;-)
[shmem]: grappa, ginebra o amaretto?
[shmem]: LanX: tried strange stuff wrt autocomplete. No luck :(
[erix]: sambuca

How do I use this? | Other CB clients
Other Users?
Others pondering the Monastery: (12)
As of 2017-03-28 12:13 GMT
Find Nodes?
    Voting Booth?
    Should Pluto Get Its Planethood Back?

    Results (330 votes). Check out past polls.