Date and Time: April 29th (Tuesday), 10:00am – 11:00am (Central Time)
Talk type: In-person | Venue: TBD
Talk Title: High-Performance Computing: An Example in Industry
Talk Overview: The use of high-performance computing – whether based on capacity, capability, or some combination
thereof – to solve technical problems has become a commonplace. Prior to the introduction of
massively parallel machines, the increase in machine speed and memory often meant using the same
mathematical approach to solution but perhaps with a more detailed model. In short, and simplistically,
wall-clock time might remain the same for solution, yet the answer was perceived to be – and
admittedly often was – more accurate as a result of the increase in detail. Here, by mathematical
approach, the term is intended to represent the linearization, discretization, numerical methods, etc.,
used to achieve an approximate solution. The advent of massively parallel machines caused, in many
cases, a reassessment of the mathematical approach to solution – not just how to use thousands of
processors effectively, but also how to enhance the approximations associated with the mathematical
approach. In some cases, it caused a reassessment of how to approach the underlying model. This talk
will take a brief tour of what are typically viewed to be driving forces for the start of computing and the
move to massively parallel machines. An example from the upstream oil and gas industry, with a
requisite brief explanation of what that is, will offer some insights into how the approach to the
underlying model mentioned above has evolved.
Speaker: Dr. John Kuzan joined ExxonMobil in 1990 as a reservoir engineer working at the Upstream Research
Company in special core analysis, which means making fluid mechanics measurements on rock
samples from petroleum reservoirs. Computational studies soon followed to create models to
mimic what he observed in the laboratory experiments and in the field. In 1995, as an extension
of those studies, John became a team leader for development of ExxonMobil’s new reservoir
simulator, which made its début in 2000. He then moved to supervise the reservoir modeling section, thus guiding the staff using the new reservoir simulator.
John subsequently managed ExxonMobil’s Breakthrough Research initiative, bringing substantial
change to its administration and management approach; that approach was still in use 25 years
later. In 2005 he began serving as the Transition Manager for ExxonMobil’s new partnership
with the Abu Dhabi National Oil Company in Zakum field. Elements of the partnership included
construction and staffing of the ExxonMobil Technology Center at Abu Dhabi, as well as defining
and initiating the work to be performed at the Center in support of Zakum field.
After this assignment, John returned to the ExxonMobil Research Company to plan and lead the
Company’s reorganization into a matrix-based research and development organization. The reorganization created an integrated project for the development of ExxonMobil’s next-
generation reservoir modeling and simulation software platform; John became the manager of that integrated project. The studies for the reorganization indicated that computational and data sciences were entering a new era and deserved renewed focus in the Research Company.
John became a founding director and ultimately managed what became the Computational and
Data Sciences Division, folding the next-generation modeling project into that Division. As the
reservoir modeling & simulation platform became usable, John moved to lead the Reservoir
Division. Similar to the move in 2000, this allowed him to guide the managers and supervisors
implementing the tool to manage ExxonMobil’s reservoirs world-wide. He retired in 2020.
John is a Chemical Engineer by training and his PhD (1986) is from the University of Illinois,
where his primary work was in experimental and computational turbulent fluid mechanics. He
attended the University of Notre Dame for his undergraduate degree (1981) on an Army
R.O.T.C. scholarship. John was on Notre Dame’s varsity track and field team (discus, shot,
hammer, and javelin) for four years; he is a member of the Monogram club.
An educational delay from entering active duty allowed John to complete his PhD; he joined the
local Army Reserve and commanded the 378th Chemical Company. Subsequently on active
duty, he was stationed at the Army’s Ballistic Research Laboratory and worked on supersonic
flow and rocket dynamics. As a team leader, he developed a research project to measure the
pressure on the exterior of artillery projectiles (with rocket-assist) while in flight. With those
measurements, the team developed a mathematical model – using the recently purchased Cray
supercomputer – that benefited the design of additional rocket-assisted projectiles. Upon
joining ExxonMobil, John continued to serve in the Army Reserve and, among many other
assignments, commanded the 450th Chemical Battalion. He retired from the Army in 2002.
John and his spouse of over 40 years, Gail, live in Houston.