Kunle Olukotun, Stanford University
The Pervasive Parallelism Laboratory (PPL)
In this talk I will give an overview of the Pervasive Parallelism Laboratory (PPL). The goal of the PPL is to make parallelism accessible to average software developers so that it can be freely used in all computationally demanding applications. The PPL pools the efforts of many leading Stanford computer scientists and electrical engineers with support from Sun Microsystems, Advanced Micro Devices, NVIDIA, IBM, Hewlett Packard and Intel. To drive the PPL research we are developing new applications in the domains of virtual worlds, personal robotics, and large-scale data informatics. These applications have the potential to exploit significant amounts of parallelism but also present significant software development challenges. The core of our research agenda is to allow domain experts to develop parallel software without becoming an expert in parallel programming. During the talk I will describe the key aspects of our approach which include domain-specific languages (DSLs), an object-oriented common parallel runtime system, and an underlying architecture that provides efficient mechanisms for communication, synchronization, and performance monitoring.