In SQL that executes aggregation, AllocSetReset is called many times and
spend a lot of cycles.
This patch saves the cycles spent by AllocSetReset.
An idea of the patch is to add a flag to AllocSetContext. This flag
shows whether AllocSetReset should work.
The effect of the patch that I measured is as follows:
o Data for test was created by 'pgbench -i -s 5'.
o Test SQL:
select b.bid, sum(a.abalance), avg(a.abalance)
from accounts a, branches b
where a.bid = b.bid
group by b.bid;
o I measured time that executed the SQL ten times.
(1)Linux(CPU: Pentium III, Compiler option: -O2)
- original: 31.310s
- patched : 28.812s
(2)Linux(CPU: Pentium 4, Compiler option: -O2)
- original: 8.953s
- patched : 7.753s
(3)Solaris(CPU: Ultra SPARC III, Compiler option: -O2)
- original: 41.8s
- patched : 38.6s
o gprof result(Linux, Compiler option: -O2 -pg -DLINUX_PROFILE)
- original
Each sample counts as 0.01 seconds.
% cumulative self self total
time seconds seconds calls s/call s/call name
9.20 3.06 3.06 38500155 0.00 0.00 AllocSetReset
8.99 6.05 2.99 27500055 0.00 0.00 slot_deform_tuple
7.40 8.51 2.46 44000000 0.00 0.00 slot_getattr
4.81 10.11 1.60 27500110 0.00 0.00 ExecEvalVar
3.64 11.32 1.21 38500143 0.00 0.00 MemoryContextReset
3.64 12.53 1.21 6007086 0.00 0.00 LWLockRelease
3.31 13.63 1.10 5500079 0.00 0.00 heapgettup
- patched
Each sample counts as 0.01 seconds.
% cumulative self self total
time seconds seconds calls s/call s/call name
8.76 2.82 2.82 27500055 0.00 0.00 slot_deform_tuple
7.73 5.31 2.49 44000000 0.00 0.00 slot_getattr
4.72 6.83 1.52 27500110 0.00 0.00 ExecEvalVar
4.32 8.22 1.39 5500011 0.00 0.00 ExecHashJoin
4.28 9.60 1.38 6007086 0.00 0.00 LWLockRelease
4.04 10.90 1.30 38500143 0.00 0.00 MemoryContextReset
3.63 12.07 1.17 5500079 0.00 0.00 heapgettup
3.04 13.05 0.98 5499989 0.00 0.00
ExecMakeFunctionResultNoSets
2.67 13.91 0.86 5500110 0.00 0.00 ExecProject
2.61 14.75 0.84 11000000 0.00 0.00 advance_transition_function
2.55 15.57 0.82 38500155 0.00 0.00 AllocSetReset
regards,
---
Atsushi Ogawa
