Proceedings of the 14th ACM SIGPLAN symposium on Principles and practice of parallel programming

Abstract

It is our great pleasure to welcome you to the 14th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP'09). Since its inception (as PPEALS) in 1988, PPoPP has served as a leading forum for research in all aspects of parallel software, including theoretical foundations, programming models, algorithms, applications, and systems software. These topics have now moved to center stage in their relevance to mainstream computing, as parallelism becomes ubiquitous with the widespread adoption of homogeneous and heterogeneous multicore processors. This year's call attracted 109 paper submissions, which is a record for PPoPP. Each paper was read by at least three members of the program committee, and additional external reviews were solicited when needed. At an in-person meeting of the program committee in October 2008, 26 papers were selected for presentation at the conference that spanned a broad range of software topics (languages, compilers, tools, runtimes) for a diversity of hardware platforms (multicore, accelerators, and high end computing). Authors of high quality submissions that could not be accepted as full papers were invited to present their work as posters. 14 of 20 accepted this invitation; 2 other posters were chosen from among 6 poster submissions. Following last year's successful co-location with the International Symposium on High-Performance Computer Architecture (HPCA) as back-to-back events, the PPoPP and HPCA conference committees decided to go one step further in 2009 and co-locate the conferences in both space and time as a single federated event. As a result, attendees of one conference have the option of attending talks at another conference, and a number of events (two keynotes, two panels, a poster session, and the conference excursion) have been scheduled as shared events. Given the significance of software-hardware co-design in addressing future research challenges in parallel systems, we hope that this coordination of PPoPP and HPCA will foster greater interaction between the two communities. In addition, our hope is that this overlap will strengthen participation in the workshops and tutorials that preceded both conferences, including the 4th ACM SIGPLAN Workshop on Transactional Computing (TRANSACT 2009).