Aart Bik's Resume

Personal Information

Education and Work History

Awards and Major Achievements

Experience

• At the Utrecht University, Aart was a Teaching Assistant for a variety of courses (Parallel Programming, Fundamental Algorithms, Distributed Databases, and Computer Architecture). During 1989 and 1990, he was member of a committee that took care of purchasing scientific books for students. At the Leiden University, Aart was a Teaching Assistant for several other courses (Operating Systems, Digital Techniques, and Compiler Construction).
• During his PhD research under supervision of Professor H.A.G. Wijshoff, Aart designed and fully implemented a special FORTRAN restructuring compiler, referred to as a sparse compiler. This source-to-source restructuring compiler automatically transforms a dense program (in which all operations on matrices are simply implemented using two-dimensional arrays) into a semantically equivalent sparse program (operating on more complicated sparse data structures), thereby reducing storage requirements and computational time of the original application. His PhD thesis "Compiler Support for Sparse Matrix Computations" received the C.J. Kok Award (outstanding thesis award) from the Leiden University.
• In 1996 and 1997, Aart was a post-doctoral researcher at the Indiana University, Bloomington, Indiana, where he conducted research in high-performance compilers for the Java programming language under supervision of Professor D.B. Gannon. This research resulted in the implementation of JAVAR (a prototype Java restructuring compiler) and JAVAB (a prototype bytecode parallelization tool). The source code of both tools, freely made available for download, has been incorporated in several other research projects.
• In January 1998, Aart started at the Microcomputer Research Laboratories of Intel Corporation in Santa Clara, California, USA, working on Java JIT compilation. This work was presented in 1999 at two tutorials during the ACM Java Grande Conference and the ACM International Conference on Supercomputing. In 1999, he transferred to the Software and Solutions Group, where he was the lead compiler architect of automatic vectorization in the Intel C++/Fortran compilers for Windows, Linux, and Mac OS. He has implemented a wide range of compiler analysis and optimization techniques, ranging from traditional data dependence analysis and loop-oriented vectorization to more multimedia-oriented alignment optimizations and vectorization of straight-line code. During this period, Aart received four divisional recognition awards and eight Intel compiler team awards. In March 2002, Aart received the Intel Achievement Award (highest company award) for making SSE easy to use through automatic vectorization. In April 2005, he was promoted to Principal Engineer. Aart published on vectorization in "The Software Vectorization Handbook" and selected chapters of "The Software Optimization Cookbook".
• In May 2007, Aart moved to Google Inc. in Mountain View, California, USA. Here he worked on Pregel, a distributed system for large-scale graph computation (viz. billions of vertices and edges), and was awarded the 2010 EngOps Citizenship Award. He joined the Graph500 steering committee in September 2011 and the Editorial Board of the International Journal of Parallel Programming in 2012. Aart co-authored the paper "Pregel: A System for Large-Scale Graph Processing", which won the 2020 ACM SIGMOD Test of Time Award. In March 2013, he joined Google[x], where he worked on the GDK (Glass Development Kit). From July 2015 to May 2018, he worked on the optimizing compiler for ART (Android Runtime). In 2017, Aart received the Google Silver Perfy Award for his work on SIMDization in the ART compiler. He presented this work at Google I/O and Linaro Connect. From May 2018 until December 2019, he worked in the Dart VM and compiler team. Since November 2019, Aart is working in the MLIR compiler team, where he revamped his old passion for vectorization and compiler support for sparse computations, while working on a new passion for machine learning!

Skills

Publications

• Aart has published extensively at international conferences and in scientific journals (over 50 publications total), and was (co-)inventor of over 10 US patents.
• Aart authored "The Software Vectorization Handbook. Applying Multimedia Extensions for Maximum Performance". ISBN 0-9743649-2-4, Intel Press, June, 2004.
• Aart co-authored "The Software Optimization Cookbook (second edition). High-Performance Recipes for the Intel Architecture". ISBN 0-9764832-1-1, Intel Press, March, 2006.
• Full list: https://www.aartbik.com/publications.php

Professional Societies and Committees

• Association for Computing Machinery (ACM).
• International Computer Games Association (ICGA).
• Member of various program committees and the editorial board of IJPP.

Personal Interests

• Aart published several Android applications on Google Play: https://play.google.com/store/apps/developer?id=Aart+Bik
• Aart implemented chess, checkers, and reversi engines, and pushed the boundaries of perft computations: https://www.aartbik.com/strategy.php
• Aart maintains a MOS Technology 6510 cross-assembler (and disassembler): https://www.aartbik.com/retro.php
• Aart is webmaster of several other websites: https://chess4android.com/, https://ilcsaratoga.org/, https://karinabik.com/.