Difference between revisions of "ArrayFire"
Jump to navigation
Jump to search
Moskalenko (talk | contribs) m (Text replacement - "#uppercase" to "uc") |
|||
| Line 38: | Line 38: | ||
--> | --> | ||
==System Variables== | ==System Variables== | ||
| − | * HPC_{{ | + | * HPC_{{uc:{{#var:app}}}}_DIR - installation directory |
<!--Configuration--> | <!--Configuration--> | ||
{{#if: {{#var: conf}}|==Configuration== | {{#if: {{#var: conf}}|==Configuration== | ||
Revision as of 21:03, 6 December 2019
Description
ArrayFire allows your code to run its fastest. It beats efforts to manually write CUDA or OpenCL kernels. It beats compiler optimizations. It beats other libraries. ArrayFire is the simply best way to accelerate your code.
Want proof? Check out our benchmarks. All benchmarks were performed on a NVIDIA Tesla C2050 with an Intel i7-950 processor (updates coming soon). The benchmarks compare ArrayFire with several popular CPU-based acceleration libraries, including Intel Math Kernel Library, Intel Integrated Performance Primitives, Eigen, and Armadillo.
Required Modules
Serial
- arrayfire
System Variables
- HPC_ARRAYFIRE_DIR - installation directory