More Details on Performance

• Base Code: JIT 3.8 and IBM Compiler 2.1 mflops

• Remove runtime checks 33.3 mflops

  • C with “Java Rules” also 33.3 mflops
  • Check each index in bound; check for null pointer

• C: Use rectangular array -- not array of pointers -- 44 mflops

• C: Use Hardware fused multiply-add -- 64 mflops

• C: Use standard compiler optimizations (associativity) 138 mflops

• Fortran: 205 megaflops

• ESSL: 253 megaflops

Previous slide

Next slide

Back to first slide

View graphic version