|
|
COSMA5 Intel Xeon CPU E5-2670
|
|
|
|
|
|
`icc` compiler version: 16.0.2 20160204
|
|
|
|
|
|
C2 Wendland Kernel
|
|
|
------------------
|
|
|
testPair (Shuffled Particles)
|
|
|
--------
|
|
|
|
|
|
Raw Times
|
|
|
---------
|
|
|
|
|
|
|CFLAGS (-no-prec-sqrt -fp-model fast=2 -qopenmp) | Sorted Cache h=1.2348 case [ticks] |h=12.5| PG Max Distance h=1.2348 case [ticks] |h=12.5|
|
|
|
|-------|-----------------------|-----------|---------------------------------------|---------|
|
|
|
|`-no-vec -no-simd -O3 -xAVX`| | | | |
|
|
|
|`-O3 -xAVX`| | | | |
|
|
|
|
|
|
Original Serial SWIFT run
|
|
|
|
|
|
|CFLAGS (-no-prec-sqrt -fp-model fast=2 -qopenmp) | h=1.2348 case [ticks] |h=12.5|
|
|
|
|-------|-----------------------|---------|
|
|
|
|`-no-vec -no-simd -O3 -xAVX`| | |
|
|
|
|
|
|
Speed Up
|
|
|
--------
|
|
|
|
|
|
|CFLAGS (-no-prec-sqrt -fp-model fast=2 -qopenmp) | Speed Up (Sorted Cache) h=1.2348 |h=12.5| Speed Up PG Max Distance h=1.2348|h=12.5|
|
|
|
|-------|-----------------------|-----|---------------------------------------|-------|
|
|
|
|`-O3 -xAVX`| | | | |
|
|
|
|
|
|
|CFLAGS (-no-prec-sqrt -fp-model fast=2 -qopenmp) | Speed Up Over Serial SWIFT (Sorted Cache) |h=12.5| Speed Up Over Serial SWIFT PG Max Distance |h=12.5|
|
|
|
|-------|-----------------------|-----|---------------------------------------|-------|
|
|
|
|`-O3 -xAVX`| | | | | |