Performance Notes

This page contains some comparisons of jRate's and RTSJ Reference Implementation (RI) performance. The benchmark used to evaluate performances was RTJPerf which is described in this paper and available for download here.

The platform used to evaluate performances had the following charateristics:

$ cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 8
model name      : Pentium III (Coppermine)
stepping        : 3
cpu MHz         : 701.602
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr sse
bogomips        : 1399.19					
and was running the TimeSys Linux/RT 3.1 GPL edition.

The set of performance results reported here are limited to:

This test shows the performance of jRate allocators versus RI allocators.

jRate allocators vs. RI allocators

This test measures the delay experienced in handling an event once it has been fired.

Event Count vs. Dispatch Delay (usec)

Dispatch Delay (usec) vs. Dispatch Delay Cumulative Frequency Distribution

This tests measures the accuracy with which the waitForNext method is implemented in RealtimeThreads.

Accuracy of waitForNext method

Last modified 15 Jul 2006 / xhtml1 / css2