nodejs / api

API WG
61 stars 24 forks source link

Stable AIB PoC - Level down performance analysis for linux #26

Open mhdawson opened 8 years ago

mhdawson commented 8 years ago

See https://github.com/nodejs/api/issues/25 for similar runs/data on windows

Steps followed:

Summary:

The differences in the benchmarks are quite small, and we probably need to do more to reduce variation on the machine and do more runs to consider the deltas valid.

Don't have times for running the tests as there is a failure in the ABI mode.

Perf Test leveldown-v8 leveldown-napi Delta
db-bench.js 72.67 sec 75 sec +3.21 %
write-random.js 193 sec 195.7 sec +1.4%
write-sorted.js 119.3 sec 117.67 sec -1.37%
tests - - -%
mhdawson commented 8 years ago

OS and CPU info for machine:

bash-4.2$ cat /etc/redhat-release
Red Hat Enterprise Linux Server release 7.2 (Maipo)
bash-4.2$ uname -a
Linux drx-hemera 3.10.0-327.18.2.el7.x86_64 #1 SMP Fri Apr 8 05:09:53 EDT 2016 x86_64 x86_64 x86_64 GNU/Linux
processor       : 119
vendor_id       : GenuineIntel
cpu family      : 6
model           : 62
model name      : Intel(R) Xeon(R) CPU E7-8880 v2 @ 2.50GHz
stepping        : 7
microcode       : 0x70d
cpu MHz         : 1712.597
cache size      : 38400 KB
physical id     : 3
siblings        : 30
core id         : 14
cpu cores       : 15
apicid          : 125
initial apicid  : 125
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm ida arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase smep erms xsaveopt
bogomips        : 4996.44
clflush size    : 64
cache_alignment : 64
address sizes   : 46 bits physical, 48 bits virtual
power management:
Believe it is 4 of these:
http://ark.intel.com/products/75257/Intel-Xeon-Processor-E7-8880-v2-37_5M-Cache-2_50-GHz
Its a big machine we have for perf tests.
mhdawson commented 8 years ago

Base benchmark times, existing leveldown implementation

Wrote 1000000 entries in 70s, 110.62MB Database size: 109M Wrote 1000000 entries in 77s, 110.62MB Database size: 110M Wrote 1000000 entries in 71s, 110.62MB Database size: 109M

Wrote 10000000 in 196s Wrote times to ./write_random_times.csv Database size: 389M Wrote 10000000 in 194s Wrote times to ./write_random_times.csv Database size: 379M Wrote 10000000 in 189s Wrote times to ./write_random_times.csv Database size: 385M

Wrote 10000000 in 121s Wrote times to ./write_sorted_times.csv Database size: 194M Wrote 10000000 in 122s Wrote times to ./write_sorted_times.csv Database size: 194M Wrote 10000000 in 115s Wrote times to ./write_sorted_times.csv Database size: 194M

Test times

real 0m7.912s user 0m2.285s sys 0m0.561s

real 0m7.819s user 0m2.238s sys 0m0.525s

real 0m7.889s user 0m2.227s sys 0m0.549s

mhdawson commented 8 years ago

Running into an error once I switch over to the napi version, will need to investigate.

mhdawson commented 8 years ago

Despite error in tests benchmarks still ran:

Wrote 1000000 entries in 76s, 110.62MB Database size: 110M Wrote 1000000 entries in 75s, 110.62MB Database size: 110M Wrote 1000000 entries in 74s, 110.62MB Database size: 110M

Wrote 10000000 in 192s Wrote times to ./write_random_times.csv Database size: 376M Wrote 10000000 in 203s Wrote times to ./write_random_times.csv Database size: 389M Wrote 10000000 in 192s Wrote times to ./write_random_times.csv Database size: 378M

Wrote 10000000 in 117s Wrote times to ./write_sorted_times.csv Database size: 194M Wrote 10000000 in 119s Wrote times to ./write_sorted_times.csv Database size: 194M Wrote 10000000 in 117s Wrote times to ./write_sorted_times.csv Database size: 194M

mhdawson commented 8 years ago

FYI - @ianwjhalliday