Slides - jtres 2012
Slides - jtres 2012 Slides - jtres 2012
Apps generate more traces, yet app traces are executed far fewer than benchmark traces• Perhaps even not enough to justify the JITC overhead Is JITC really useful for App performance?Virtual Machine & Optimization Lab32
Loading time only 1.21.110.90.80.7AngrybirdsDoodleJump Seesmic Twitter Astro FileManagerGoogle SkyMapGeomeanInterpreterJITCApp performance goes down when we turn on JIT compiler Virtual Machine & Optimization Lab33
- Page 4: VM for executing Java in Android pl
- Page 11 and 12: Java Source Codepublic static int f
- Page 13 and 14: 2.521.510.50Chess kXML Parallel PNG
- Page 15 and 16: 2.521.510.50Chess kXML Parallel PNG
- Page 17 and 18: 1.81.61.41.210.80.60.40.20Chess kXM
- Page 19 and 20: How many times a Dalvik bytecode is
- Page 21 and 22: 86.00%75.00%654.00%43.00%32.00%211.
- Page 23 and 24: Low code quality due to short trace
- Page 25 and 26: 100%80%60%40%20%0%Native Native app
- Page 27 and 28: Garbage collection (GC) portion is
- Page 29 and 30: Numbers are log scale 1000000010000
- Page 31: 500450400350300250200150100500App t
Loading time only 1.21.110.90.80.7AngrybirdsDoodleJump Seesmic Twitter Astro FileManagerGoogle SkyMapGeomeanInterpreterJITCApp performance goes down when we turn on JIT compiler Virtual Machine & Optimization Lab33