Skip to Main Content

Dr. Kunle Olukotun: Inventor, Professor and Pioneer of Multi Core Scalar Processors

This profile on Dr. Kunle Olukotun, an inventor and professor of Electrical Engineering and Computer Science at Stanford University, is the twelfth post in a month-long series of profiles on Black STEM innovators in honor of Black History Month. Today’s post also celebrates Engineers’ Week, by focusing on Dr. Olukotun. 

Dr. Kunle Olukotun is well known in the engineering industry for his substantial contributions to the field in the form of tactile inventions and in-depth research. Before beginning his professional work, Dr. Olukotun worked vigilantly as a student to earn a B.S degree from Calvin College in Michigan, then an M.S. and PhD in Computer Science and Engineering from the University of Michigan in 1987 and 1991, respectively1.

The same year he earned his doctorate, Dr. Olukotun joined the faculty of Stanford University as the Cadence Design Systems Professor in the School of Engineering and Professor of Electrical Engineering and Computer Science1,2

While at Stanford, Dr. Olukotun became the leader of the Stanford Hydra chip multiprocessor (CMP) research project. The project, described as “a promising way to build a small-scale MP-on-a-chip with a fairly simple design while maintaining excellent performance on a wide variety of applications3” allowed for the development of “one of the first chip multiprocessors with support for thread-level speculation (TLS)2.” 

With this groundbreaking experience and prolific research under his belt, Dr. Olukotun founded Afara Websystems with the goal of developing “high-throughput, low-power multicore processors for server systems”4. Afara was acquired by Sun Microsystems which used Dr. Olukotun’s previous technological innovations to create the UltraSPARC T1 microprocessor, more commonly known as "Niagara", a multithreading, multicore CPU1

The work Dr. Olukotun contributed to the industry was focused on the idea of parrallelism in computing. Hydra Chip MultiProcessor (CMP) was a pioneering project in multicore processors which involve joining multiple individual processors on a single chip. He and his colleagues believed that multi core scalar processors, which orchestrate work between multiple processors that only execute one instruction at a time, would be more effective than the predominately used superscalar processing in which a single processor has different execution units allowing multiple instruction executions simultaneously. This was the ground work for modern personal computers and phones that currently utilize multiple core processors.

In his current work at Stanford, Dr. Olukotun serves as the director of the Stanford Pervasive Parallelism Lab (PPL), which aims “to proliferate the use of heterogeneous parallelism in all application areas using domain-specific languages”4

In addition to his lauded research contributions and innovative technological inventions, Olukotun is a fellow of both the Institute of Electrical and Electronics Engineers (IEEE) and of the Association for Computing Machinery (ACM), and a member of the Data Analytics for What’s Next (DAWN) Lab1,4.

Currently, Dr. Olukotun is “focusing on the design of domain-specific accelerators for machine learning and data analytics using parallel patterns” at the DAWN Lab in addition to teaching the next generation of engineering and computer science professionals at Stanford.

For a list of Dr. Olukotun’s research, visit: https://scholar.google.com/citations?user=IzXDyR8AAAAJ&hl=en

References:

  1. Stanford Profiles. (2020). Kunle Olukotun. Retrieved from https://profiles.stanford.edu/20335
  2. Kunle Olukotun. (2012). Bio. Retrieved from http://arsenalfc.stanford.edu/kunle/.
  3. Stanford University. (1998, February). Considerations in the Design of Hydra: A multiprocessor-on-a-chip Microarchitecture. Retrieved from http://i.stanford.edu/pub/cstr/reports/csl/tr/98/749/CSL-TR-98-749.pdf.
  4. IEEE Computer Society. Kunle Olukotun. Retrieved from https://www.computer.org/profiles/kunle-olukotun.