Stabilize Project Benchmark Performance

The data situation of a stable project after six months of online (intercept a share)

Data for internal project testing
XXXX did a version of performance testing-benchmark test
XXXXXX Application Version 1.1.2 1212 Resource version
App Package: Cc.xxxxx.xxxxx.qihoo
Operation Mobile: Xiaomi mi4w Operating system: 4.4.4 (Stable)
In-game running for 30 minutes
CPU avg:27.52% slightly higher
CPU max:53.23% slightly higher is instantaneous control within 3 seconds
Mem AVG:150.961MB 20MB higher than the previous 2 versions of the package
Mem MAX:194.94MB Memory is flat
OS Distribution MEM:64MB automatic assignment
Ah 2.3%/h OK
Hover on the chat interface CPU will grow 5~7%, non-instantaneous behavior after destroying the interface to return to normal, multiple operations will also grow, but moving to other interfaces will return to normal, some pages loaded with too much information caused. (also on more than 3 layers, login to collect rewards are present)


NO ERRORS FOUND:
0 unique, 0 total unaddressable access (es)
0 unique, 0 Total uninitialized access (es)
0 unique, 0 total Invalid heap argument (s)
0 unique, 0 Total GDI usage error (s)
0 unique, 0 Total handle leak (s)
0 Unique, 0 Total warning (s)
0 unique, 0 total, 0 byte (s) of leak (s)
0 unique, 0 total, 0 byte (s) of possible leak (s)

Add:

Performance Test Benchmarks

Performance tests for non-functional points are visible for each version of the non-functional point

Example of performance testing for main scenes and fixed operations 2

The above 2 types are recorded parameters, each milestone version of the difference between the first, if each version of the difference is large, the beginning of analysis is a new resource problem or code problems, whether there is code is overwritten, etc., the keyword is the difference.

When the performance test starts, it is necessary to do so when the debug version issue is over.

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

Stabilize Project Benchmark Performance