Development Tools

Titan will be configured with a broad set of tools to facilitate GPU acceleration of applications both existing and new. These tools can be broken down into three categories based on their implementation methodology: GPU-accelerated libraries, accelerator compiler directives, and low-level GPU languages. Each of these three methods has pros and cons that must be weighed for each program and are not mutually exclusive. In addition to these tools, Titan will support a wide variety of performance and debugging tools to ensure that however you choose to implement GPU acceleration into your program, it is being done so efficiently.

Accelerator compiler directives allow the compiler, guided by the programmer, to take care of low-level accelerator work. One of the main benefits of a directives-based approach is an easier and faster transition of existing code compared to low-level GPU languages. Additional benefits include performance enhancements that are transparent to the end developer and greater portability between current and future many-core architectures.

Due to the performance benefits that come with GPU computing, many scientific libraries are now offering accelerated versions. If your program contains BLAS or LAPACK function calls, GPU-accelerated versions may be available. Magma, CULA, cuBLAS, and cuSPARSE libraries provide optimized GPU linear algebra routines that require only minor changes to existing code. These libraries require little understanding of the underlying GPU hardware, and performance enhancements are transparent to the end developer.

For more general libraries, such as Trilinos and PETSc, you will want to visit the software development site to checkout the current status of GPU integration. For Trilinos please see the latest documentation. PETSc has a web page containing the latest GPU integration information.

For complete control over the GPU Titan will support C for CUDA, PGI’s CUDA Fortran, and OpenCL. These languages and language extensions, while allowing explicit control, are generally more cumbersome than directive-based approaches and must be maintained to stay up to date with the latest performance guidelines. Substantial code structure changes may be needed and an in-depth knowledge of the underlying hardware is often necessary for best performance.

To ensure efficient use of Titan a full suite of performance and analysis tools will be offered. These tools offer a wide variety of support from static code analysis to full runtime heterogeneous tracing.