• Program, papers, and talks are now available.
• Thanks to all who attended and participated in the workshop.

Parallel and distributed computing platforms provide massive levels of concurrency that can be exploited for better performance. However orchestrating computation over such platforms requires a lot of hard work, and is fraught with programming pitfalls that may lead to runtime errors and anomalies that are hard to debug. Stream programming is gaining increasing attention because it eases the burden associated with designing and implementing complex, large-scale, and scalable applications for parallel and distributed computing platforms. In a stream programming paradigm, programmers focus on designing their applications. Specifically, they describe their computation programmatically and algorithmically, and do not commit to specific implementation details related to scheduling, buffering, synchronization, or the underlying data transport mechanisms in their target platforms. This programming practice leads to code that is easy to maintain, modify and port. This workshop will look at how streaming applications are mapped to various architectures including uniprocessors, homogeneous and heterogeneous multicores, FPGA-accelerated systems, and/or novel architecture that use optical interconnect.

Streaming systems can be found in many tiers of the computing industry. The workshop focuses on the following types of stream computing.

• Web and Enterprise: streaming in this domain tends to offer massively scalable high-performance for rapidly analyzing data as it streams in from multiple sources. It is designed and engineered for the kind of large scale stream computing that abounds in the Web and enterprise domains, with computing power spanning tens of thousands of distributed processors.
• Multicore: applications in this domain are designed for scalable stream computing on DSPs, embedded, and desktop processors, where parallelism is currently embodied in multicore processors and small clusters with with 10s-100s of processors.

The workshop provides a venue to bring together researchers and practitioners working on streaming systems for Web, Enterprise, and Multicores. Position statements are solicited in areas including but not limited to the following:

• Real streaming scenarios and applications from Web and Enterprise
• Streaming applications for Multicore
• Types of parallelism exposed by streaming applications
• Application design methodologies
• Programming language principles and implementation
• Opportunities for stream-aware compiler optimizations
• Scheduling and orchestration of computation
• Streaming runtime systems for heterogeneous architectures
• Architectural design space for streaming

Please send submissions by the deadline to streamingsystems@gmail.com. Include a complete list of authors, their affiliations and contact information (e.g., address, telephone number, and email address), and please identify the corresponding author with your submission.