VIVEKFEST 2024

The Vivek Sarkar Festschrift Symposium is being planned during 20-24th Oct, 2024 in Pasadena, California at the SPLASH’24. This day-long symposium will be to celebrate the scholarship and teaching of Prof. Vivek Sarkar, the Chair of the School of Computer Science at Georgia Tech and the Stephen Fleming Chair for Telecommunications in the College of Computing.

The theme of the symposium is “Principles and Practices of Building Parallel Software” reflecting Vivek’s profound dedication to exploring the multifaceted realm of parallel computing software, encompassing programming languages, compilers, runtime systems, and debugging and verification systems tailored for high-performance computing environments. Vivek started his journey in the industry before bringing those valuable industry practices to academia. Specifically, his research projects at IBM include the PTRAN automatic parallelization system, the ASTI optimizer for IBM’s XL product compilers, the open-source Jikes Research Virtual Machine for the Java language, and the X10 programming language developed in the DARPA HPCS program. At academia he has been instrumental in pushing the forefront of system software in the Habanero Extreme Scale Software Research project. Vivek is an ACM Fellow, an IEEE Fellow, a former member of the IBM Academy of Technology, and a recipient of the ACM-IEEE CS Ken Kennedy Award. He served on the Board of Directors of the Computing Research Association (CRA) during 2015-2022, and is a founding co-chair of the CRA-Industry committee since 2021. He has been a member of the US Department of Energy’s Advanced Scientific Computing Advisory Committee (ASCAC) since 2009, and a member of the ACM Council since 2022.

Vivek is interested in the performance and efficiency of parallel computing systems, which are crucial for managing large-scale computational tasks. Over the years, his work has been to develop innovative compiler techniques, runtime systems, and programming models that can effectively harness the power of multicore, manycore, and heterogeneous processors. He has particularly focused on scalable and productive programming methodologies, aiming to bridge the gap between the high-level abstraction preferred by programmers and the low-level efficiency required for high-performance computing tasks. Through his research, Vivek endeavored to address the challenges of concurrency, synchronization, and memory management in parallel systems, thereby contributing to advancements in computing technologies that are increasingly becoming integral to various scientific, engineering, and commercial applications.

The symposium is dedicated to Vivek’s lifelong ideas and influences, and some of Vivek’s closest collaborators and most prominent colleagues will deliver talks during the event. They will also be invited to contribute a paper to a so-called Festschrift, which will be published either by ACM or by Springer in their Lecture Notes in Computer Science (LNCS) series (tentative). The term “Festschrift” is borrowed from German, and could be translated as celebration publication or celebratory (piece of) writing (literally ‘party-writing’; cognate with ‘feast-script’). The symposium is open to all interested attendees.

Call for Papers

The scope of the Vivek Sarkar Festschrift Symposium aims to cover a wide range of subjects within parallel computing, such as programming models, program analysis and transformations, runtime systems and execution models, and application-oriented software tools and libraries. The domain of interest will be any system in which parallel and heterogeneous processing plays a key role. In addition, we will encourage contributions involving the use of trained AI methods, such as leveraging GPT for code generation, program analysis and transformation to empower performance and productivity of large-scale computations.

Specific topics of interest for include:

• Compilers for parallel and heterogeneous computing

• Static, dynamic, and adaptive optimization of parallel programs

• Parallel programming models and languages

• Formal analysis and verification of parallel programs

• Parallel runtime systems and libraries

• Support for fine-grain parallel processing

• Performance and analysis and debugging tools for concurrency

• Parallel algorithms and concurrent data structures

• Parallel applications in Big Data, Machine Learning, Graphs, and Biology

• Strong scaling and issues involving extreme-scale computing systems

• Parallel programming and compiling for heterogeneous systems

• Use of AI/ML in code generation, program optimization, and parallel programming

We encourage personal engagement and technical conversations inspired by the materials presented live and in-person at the symposium. It is also expected that at least one author from each paper will need to be present and discuss the work. The symposium will welcome both invited talks and research papers.

Submitted regular papers should be previously unpublished work no longer than 10 pages (excluding bibliography) in a form compliant with the LNCS author guidelines. Short papers may represent “work in progress” and follow the same LNCS rules but with a maximum of 6 pages. Invited talks may be one-pager. Accepted papers will be made available online as a proceeding.