On Fri, 25 Nov 2005, Tom Lane wrote:
>
> Is that "modern machine" a Xeon by any chance?
>
$#cat /proc/cpuinfo | grep "model name"
model name : Intel(R) Pentium(R) 4 CPU 2.40GHz
I can find a 4way Xeon (but it is shared by many users):
/h/164/zhouqq#cat /proc/cpuinfo |grep "model name"
model name : Intel(R) Xeon(TM) CPU 2.40GHz
model name : Intel(R) Xeon(TM) CPU 2.40GHz
model name : Intel(R) Xeon(TM) CPU 2.40GHz
model name : Intel(R) Xeon(TM) CPU 2.40GHz
$/pgsql/src/backend/storage/lmgr#./a.out
Spinlock pair(2648542) duration: 161.785 ms
$/pgsql/src/backend/storage/lmgr#./a.out
Spinlock pair(2648542) duration: 160.661 ms
$/pgsql/src/backend/storage/lmgr#./a.out
Spinlock pair(2648542) duration: 155.505 ms
>
> it now occurs to me that you could still cherry-pick non-corrupt tuples
> with TID-scan queries, so this objection shouldn't be considered fatal.
>
It is great that it is not that difficult to do it! What's more, the
parallel scan will be easier to implement based on page level scan
operators.
Regards,
Qingqing