stalkerg/bufmgr_perf_top Secret

## bufmgr_perf_top
Cpu(s): 71.0%us, 18.3%sy,  0.0%ni, 10.7%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st

[root@localhost src]# uptime
 14:39:28 up 7 days, 19:20,  9 users,  load average: 106.61, 56.77, 44.59

Samples: 33M of event 'cycles', Event count (approx.): 4079875627752, UID: vadv
   4.36%  postgres            [.] GetSnapshotData
   2.88%  postgres            [.] hash_search_with_hash_value
   2.74%  postgres            [.] _bt_compare
   2.64%  postgres            [.] AllocSetAlloc
   2.32%  postgres            [.] LWLockAcquire
   2.27%  postgres            [.] SearchCatCache
   2.23%  postgres            [.] base_yyparse
   1.79%  [kernel]            [k] _spin_lock
   1.24%  postgres            [.] PinBuffer
   1.02%  postgres            [.] LWLockRelease
   0.91%  [kernel]            [k] page_fault
   0.91%  postgres            [.] core_yylex
   0.84%  [kernel]            [k] fget_light
   0.80%  libc-2.12.so        [.] __strcmp_sse42
   0.75%  postgres            [.] palloc
   0.73%  libc-2.12.so        [.] _int_malloc
   0.67%  libc-2.12.so        [.] vfprintf
   0.64%  postgres            [.] expression_tree_walker
   0.59%  postgres            [.] MemoryContextAllocZeroAligned
   0.58%  postgres            [.] nocachegetattr
   0.54%  libc-2.12.so        [.] __strlen_sse42
   0.52%  postgres            [.] grouping_planner
   0.50%  libc-2.12.so        [.] _int_free
   0.50%  [kernel]            [k] avc_has_perm_noaudit

## lscpu
[root@localhost src]# lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                144
On-line CPU(s) list:   0-143
Thread(s) per core:    2
Core(s) per socket:    18
Socket(s):             4
NUMA node(s):          4
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 63
Stepping:              4
CPU MHz:               2494.005
BogoMIPS:              4987.28
Virtualization:        VT-x
L1d cache:             32K
L1i cache:             32K
L2 cache:              256K
L3 cache:              46080K
NUMA node0 CPU(s):     0,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140
NUMA node1 CPU(s):     1,5,9,13,17,21,25,29,33,37,41,45,49,53,57,61,65,69,73,77,81,85,89,93,97,101,105,109,113,117,121,125,129,133,137,141
NUMA node2 CPU(s):     2,6,10,14,18,22,26,30,34,38,42,46,50,54,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118,122,126,130,134,138,142
NUMA node3 CPU(s):     3,7,11,15,19,23,27,31,35,39,43,47,51,55,59,63,67,71,75,79,83,87,91,95,99,103,107,111,115,119,123,127,131,135,139,143

## pgbench_bufmgr_S_scale_1000
shared_buffers = 100GB
max_clients = 500

pgbench -S  -T 600 -j 50 -c 300 pgbench
starting vacuum...end.
transaction type: SELECT only
scaling factor: 1000
query mode: simple
number of clients: 300
number of threads: 50
duration: 600 s
number of transactions actually processed: 579523167
latency average: 0.311 ms
tps = 965821.878404 (including connections establishing)
tps = 965894.829109 (excluding connections establishing)

## pgbench_vanila_S_scale_1000
shared_buffers = 100GB
max_clients = 500

pgbench -S  -T 600 -j 50 -c 300 pgbench
starting vacuum...end.
transaction type: SELECT only
scaling factor: 1000
query mode: simple
number of clients: 300
number of threads: 50
duration: 600 s
number of transactions actually processed: 252070436
latency average: 0.714 ms
tps = 417523.148047 (including connections establishing)
tps = 417570.851235 (excluding connections establishing)

## vanila_perf_top
[root@localhost src]# uptime
 14:37:29 up 7 days, 19:18,  9 users,  load average: 0.80, 11.74, 29.82


Samples: 34M of event 'cycles', Event count (approx.): 4066145469067, UID: vadv
  33.48%  postgres            [.] s_lock
   2.51%  postgres            [.] GetSnapshotData
   1.82%  postgres            [.] PinBuffer
   1.64%  [kernel]            [k] _spin_lock
   1.63%  postgres            [.] hash_search_with_hash_value
   1.51%  postgres            [.] _bt_compare
   1.44%  postgres            [.] AllocSetAlloc
   1.35%  postgres            [.] base_yyparse
   1.31%  postgres            [.] SearchCatCache
   1.25%  postgres            [.] LWLockAcquire
   0.80%  postgres            [.] LWLockRelease
   0.70%  postgres            [.] UnpinBuffer.clone.0
   0.67%  [kernel]            [k] fget_light
   0.56%  postgres            [.] BufferGetBlockNumber
   0.53%  postgres            [.] ReadBuffer_common
   0.50%  [kernel]            [k] page_fault
   0.50%  libc-2.12.so        [.] _int_malloc
   0.49%  postgres            [.] core_yylex
	Cpu(s): 71.0%us, 18.3%sy, 0.0%ni, 10.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st

	[root@localhost src]# uptime
	14:39:28 up 7 days, 19:20, 9 users, load average: 106.61, 56.77, 44.59

	Samples: 33M of event 'cycles', Event count (approx.): 4079875627752, UID: vadv
	4.36% postgres [.] GetSnapshotData
	2.88% postgres [.] hash_search_with_hash_value
	2.74% postgres [.] _bt_compare
	2.64% postgres [.] AllocSetAlloc
	2.32% postgres [.] LWLockAcquire
	2.27% postgres [.] SearchCatCache
	2.23% postgres [.] base_yyparse
	1.79% [kernel] [k] _spin_lock
	1.24% postgres [.] PinBuffer
	1.02% postgres [.] LWLockRelease
	0.91% [kernel] [k] page_fault
	0.91% postgres [.] core_yylex
	0.84% [kernel] [k] fget_light
	0.80% libc-2.12.so [.] __strcmp_sse42
	0.75% postgres [.] palloc
	0.73% libc-2.12.so [.] _int_malloc
	0.67% libc-2.12.so [.] vfprintf
	0.64% postgres [.] expression_tree_walker
	0.59% postgres [.] MemoryContextAllocZeroAligned
	0.58% postgres [.] nocachegetattr
	0.54% libc-2.12.so [.] __strlen_sse42
	0.52% postgres [.] grouping_planner
	0.50% libc-2.12.so [.] _int_free
	0.50% [kernel] [k] avc_has_perm_noaudit
	[root@localhost src]# lscpu
	Architecture: x86_64
	CPU op-mode(s): 32-bit, 64-bit
	Byte Order: Little Endian
	CPU(s): 144
	On-line CPU(s) list: 0-143
	Thread(s) per core: 2
	Core(s) per socket: 18
	Socket(s): 4
	NUMA node(s): 4
	Vendor ID: GenuineIntel
	CPU family: 6
	Model: 63
	Stepping: 4
	CPU MHz: 2494.005
	BogoMIPS: 4987.28
	Virtualization: VT-x
	L1d cache: 32K
	L1i cache: 32K
	L2 cache: 256K
	L3 cache: 46080K
	NUMA node0 CPU(s): 0,4,8,12,16,20,24,28,32,36,40,44,48,52,56,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140
	NUMA node1 CPU(s): 1,5,9,13,17,21,25,29,33,37,41,45,49,53,57,61,65,69,73,77,81,85,89,93,97,101,105,109,113,117,121,125,129,133,137,141
	NUMA node2 CPU(s): 2,6,10,14,18,22,26,30,34,38,42,46,50,54,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118,122,126,130,134,138,142
	NUMA node3 CPU(s): 3,7,11,15,19,23,27,31,35,39,43,47,51,55,59,63,67,71,75,79,83,87,91,95,99,103,107,111,115,119,123,127,131,135,139,143
	shared_buffers = 100GB
	max_clients = 500

	pgbench -S -T 600 -j 50 -c 300 pgbench
	starting vacuum...end.
	transaction type: SELECT only
	scaling factor: 1000
	query mode: simple
	number of clients: 300
	number of threads: 50
	duration: 600 s
	number of transactions actually processed: 579523167
	latency average: 0.311 ms
	tps = 965821.878404 (including connections establishing)
	tps = 965894.829109 (excluding connections establishing)
	[root@localhost src]# uptime
	14:37:29 up 7 days, 19:18, 9 users, load average: 0.80, 11.74, 29.82


	Samples: 34M of event 'cycles', Event count (approx.): 4066145469067, UID: vadv
	33.48% postgres [.] s_lock
	2.51% postgres [.] GetSnapshotData
	1.82% postgres [.] PinBuffer
	1.64% [kernel] [k] _spin_lock
	1.63% postgres [.] hash_search_with_hash_value
	1.51% postgres [.] _bt_compare
	1.44% postgres [.] AllocSetAlloc
	1.35% postgres [.] base_yyparse
	1.31% postgres [.] SearchCatCache
	1.25% postgres [.] LWLockAcquire
	0.80% postgres [.] LWLockRelease
	0.70% postgres [.] UnpinBuffer.clone.0
	0.67% [kernel] [k] fget_light
	0.56% postgres [.] BufferGetBlockNumber
	0.53% postgres [.] ReadBuffer_common
	0.50% [kernel] [k] page_fault
	0.50% libc-2.12.so [.] _int_malloc
	0.49% postgres [.] core_yylex