Personal tools
You are here: Home Events Title: Compiler-Assisted Test Acceleration on GPUs (for Embedded Software)

Title: Compiler-Assisted Test Acceleration on GPUs (for Embedded Software)

Speaker: Vanya Yaneva


Embedded software is found everywhere from our highly visible mobile devices to the confines of our car in the form of smart sensors.  Embedded software companies are under huge pressure to produce safe applications that limit risks, and testing is absolutely critical to alleviate concerns regarding safety and user privacy.  This requires using large test suites throughout the development process, increasing time-to-market and ultimately hindering competitiveness.

Speeding up test execution is, therefore, of paramount importance for embedded software developers.  This is traditionally achieved by running, in parallel, multiple tests on large-scale clusters of computers.  However, this approach is costly in terms of infrastructure maintenance and energy consumed, and is at times inconvenient as developers have to wait for their tests to be scheduled on a shared resource.

I look at exploiting GPUs (Graphics Processing Units) for running embedded software testing.  GPUs are readily available in most computers and offer tremendous amounts of parallelism, making them an ideal target for embedded software testing.  However, they use specialist programming models, which limits their scope and makes them notoriously difficult to program.  To mitigate these issues, I propose a compiler-assisted approach which automatically compiles the C program into GPU kernels and executes their tests in parallel on the GPU threads.  Current evaluation across nine programs from an industry standard embedded benchmark suite achieves an average speedup of 16x when compared to CPU execution.

In this talk, I will present this approach, together with current evaluation results and some ideas for future work. Papers (both with coauthors Ajitha Rajan and Christophe Dubach):

Compiler-Assisted Test Acceleration on GPUs for Embedded Software (ISSTA'17)
ParTeCL: Parallel Testing Using OpenCL (ISSTA'17 tools demo)


Document Actions