Ateji PX for Java: parallel programming made simple
Boost application performance and leverage hardware assets
by using all available processor cores
The number of cores in desktop computers and servers is expected to double every year. While the OS is able to run different applications on different cores, the only way for an application to benefit from a performance increase is to parallelize the code at the level of the application.
Ateji PX makes this parallelization process simple and compatible with existing Java code.
Sequential Code: without
Ateji PX the application can only use one core |
Parallel Code: with
Ateji PX the application is able to use all available cores |
![]() |
![]() |
![]() |
![]() |
Learn more | See our online demo |
As easy as '||'
12.5x speed-up on a 16-core server
The Ateji white-paper
"Matrix multiplication with Ateji PX" documents the case of the standard
benchmark for data parallelism, where a 12.5x speed-up was observed on a
16-core server.
No need to be specialized in multithreading, all what is required is the addition of a single
|| operator in the source code: for ||(int i : I) { for(int j : J) { for(int k : K) { C[i][j] += A[i][k] * B[k][j]; } } }
Key features
Ateji PX is built upon a unique innovative language-integration technology that makes parallel programming:- Simple and intuitive, accessible to all software developers
- Easy to learn and adopt
- Efficient, making the best use of multi-core hardware
- Compatible with existing code, tools and practice
- Based on standards
- Ready for future-generation hardware
Customer Benefits
- Leverage the performance of your existing multi-core chips
- Lower development cost
- Improve time-to-market
- Reduce maintainance costs
- Easy training and staffing
- Few changes in the development process

