Re: POC: Cache data in GetSnapshotData()

On Wed, Mar 16, 2016 at 10:59 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:

> On Wed, Mar 16, 2016 at 4:40 AM, Mithun Cy <mithun(dot)cy(at)enterprisedb(dot)com>
> wrote:
>
>>
>> On Thu, Mar 3, 2016 at 6:20 AM, Mithun Cy <mithun(dot)cy(at)enterprisedb(dot)com>
>> wrote:
>> > I will continue to benchmark above tests with much wider range of
>> clients.
>>
>> Latest Benchmarking shows following results for unlogged tables.
>>
>> clients BASE ONLY CLOG CHANGES % Increase CLOG CHANGES + SAVE SNAPSHOT %
>> Increase
>> 1 1198.326337 1328.069656 10.8270439357 1234.078342 2.9834948875
>> 32 37455.181727 38295.250519 2.2428640131 41023.126293 9.5259037641
>> 64 48838.016451 50675.845885 3.7631123611 51662.814319 5.7840143259
>> 88 36878.187766 53173.577363 44.1870671639 56025.454917 51.9203038731
>> 128 35901.537773 52026.024098 44.9130798434 53864.486733 50.0339263281
>> 256 28130.354402 46793.134156 66.3439197647 46817.04602 66.4289235427
>>
>
> Whoa. At 64 clients, we're hardly getting any benefit, but then by 88
> clients, we're getting a huge benefit. I wonder why there's that sharp
> change there.
>
>
If you see, for the Base readings, there is a performance increase up till
64 clients and then there is a fall at 88 clients, which to me indicates
that it hits very high-contention around CLogControlLock at 88 clients
which CLog patch is able to control to a great degree (if required, I think
the same can be verified by LWLock stats data). One reason for hitting
contention at 88 clients is that this machine seems to have 64-cores (it
has 8 sockets and 8 Core(s) per socket) as per below information of lscpu
command.

Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 128
On-line CPU(s) list: 0-127
Thread(s) per core: 2
Core(s) per socket: 8
Socket(s): 8
NUMA node(s): 8
Vendor ID: GenuineIntel
CPU family: 6
Model: 47
Model name: Intel(R) Xeon(R) CPU E7- 8830 @ 2.13GHz
Stepping: 2
CPU MHz: 1064.000
BogoMIPS: 4266.62
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 24576K
NUMA node0 CPU(s): 0,65-71,96-103
NUMA node1 CPU(s): 72-79,104-111
NUMA node2 CPU(s): 80-87,112-119
NUMA node3 CPU(s): 88-95,120-127
NUMA node4 CPU(s): 1-8,33-40
NUMA node5 CPU(s): 9-16,41-48
NUMA node6 CPU(s): 17-24,49-56
NUMA node7 CPU(s): 25-32,57-64

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com