Data parallel simulation involves simulating the behavior of a circuit over a number of test sequences simultaneously. Compared to other parallel simulation techniques, data parallel simulation requires less overhead for synchronization and communication, and it permits higher degrees of parallelism. We have implemented two data parallel versions of the switch-level simulator COSMOS. The rst ru...

There are few, if any, men who would qualify for the position of Playboy bunny, and there are few, if any, women who would qualify for the position of male lead in a motion picture. However, there are many other occupations in which gender may be irrelevant. People in these jobs include nurses, truck drivers, lawyers, secretaries, and economists. Why is it, then, that these jobs are each charac...

Here we report on the performance of GROMACS 4.6 on the SuperMUC cluster at the Leibniz Rechenzentrum in Garching. We carried out benchmarks with three biomolecular systems consisting of eighty thousand to twelve million atoms in a strong scaling test each. The twelve million atom simulation system reached a performance of 49 nanoseconds per day on 32,768 cores.

Debugging is a resource-consuming activity of software development. Some bugs are deeply rooted in the domain logic but others are independent of the specificity of the application being debugged. The latter are “crowd-bugs”: unexpected and incorrect output or behavior resulting from a common and intuitive usage of an API. On the contrary, project-specific bugs are related to the misunderstandi...

Concurrency related bugs are hard to detect and even harder to reproduce due to their time dependent nature. One big factor in non-determinism of multi-threaded applications is process scheduler of an operating system. In this work, we presented results of our study to understand the behavior of some of the selected concurrency bugs under different OS schedulers. The generalization of this stud...

Privacy in social computing systems has become a major concern. End-users of such systems find it increasingly hard to understand complex privacy settings. As software evolves over time, this might introduce bugs that breach users’ privacy. Further, there might be system-wide policy changes that could change users’ settings to be more or less private than before. We present a novel technique th...

Randomly mutating well-formed program inputs or simply fuzzing, is a highly effective and widely used strategy to find bugs in software. Other than showing fuzzers find bugs, there has been little systematic effort in understanding the science of how to fuzz properly. In this paper, we focus on how to mathematically formulate and reason about one critical aspect in fuzzing: how best to pick see...

In software development, fixing bugs is an important task that is time consuming and cost-sensitive. While many approaches have been proposed to automatically detect and patch software code, the strategies are limited to a set of identified bugs that were thoroughly studied to define their properties. They thus manage to cover a niche of faults such as infinite loops. We build on the assumption...