<div dir="ltr"><div class="gmail_extra"><br /><div class="gmail_quote">On Mon, Mar 17, 2014 at 7:01 PM, Jim Nasby <span
dir="ltr"><<ahref="mailto:jim@nasby.net" target="_blank">jim@nasby.net</a>></span> wrote:<br /><blockquote
class="gmail_quote"style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="a3s" id=":1nx"
style="overflow:hidden">Evenbetter would be if the planner could estimate how bad a plan will become if we made
assumptionsthat turn out to be wrong.<br /></div></blockquote></div><br /></div><div class="gmail_extra">That's
preciselywhat risk estimation was about.<br /><br /></div>Something like<br /><div class="gmail_extra"><br /></div><div
class="gmail_extra">SELECT* FROM wherever WHEN id > something LIMIT COST 10000;<br /><br /></div><div
class="gmail_extra">Wouldforbid a sequential scan *if* the table is big enough to suspect the plan might take that
much,or a nested loop *if* the planner cannot *prove* it will be faster than that.<br /><br /></div><div
class="gmail_extra">Idon't believe the limit unit is obscure at all (page fetches being a nice measuring stick), but
whatis, is what do you do when no plan fits the limits.<br /><br /></div></div>