Table of Contents
-
FREE
REGISTRATION
HERE!
-
Become a sponsor
HERE!
-
New to supercomputing? Click
here.
-
KEYNOTE:
Michael Norman,
San Diego Supercomputer Center,
University of California San Diego
-
PLENARY:
Dr. Robert
M. Panoff,
Shodor Education Foundation
-
PLENARY:
Dan Stanzione,
Texas Advanced Computing Center,
University of Texas at Austin
-
PLENARY:
Henry Neeman,
University of Oklahoma
-
Daniel Andresen,
Kansas State University
-
Shady Boukhary,
Midwestern State University
-
Keith Brewster,
University of Oklahoma
-
Eduardo
Colmenares,
Midwestern State University
-
Brady Deetz,
Laureate Institute for Brain Research
-
Kyle Hutson,
Kansas State University
-
Adnan Khaleel,
Dell EMC
-
Mark Laufersweiler,
University of Oklahoma
-
BJ Lougee,
Federal Reserve Bank of Kansas City
-
George Louthan,
University of Oklahoma
-
Chongle Pan,
University of Oklahoma
-
Dimitrios
Papavassiliou,
University of Oklahoma
-
Robert Peglar,
Formulus Black
PLENARY
SPEAKERS
Director
San
Diego Supercomputer Center
University
of California San Diego
Topic:
"Tales from the Long Tail of HPC:
The Comet Experiment"
Slides:
available after the Symposium
Talk Abstract
Coming soon
Biography
Dr. Michael L. Norman
was named Interim Director
of the
San
Diego Supercomputer Center
(SDSC)
at the
University
of California San Diego
(UCSD)
in June 2009
and was appointed to the position of
Director in September 2010.
Dr. Norman
is a distinguished professor of
Physics
at UCSD
and a globally recognized astrophysicist.
Dr. Norman is a pioneer in
using advanced computational methods
to explore the universe and its beginnings.
In this capacity,
he has directed the
Laboratory for Computational Astrophysics
—
a collaborative effort between UCSD and SDSC,
resulting in the
Enzo
community code for astrophysics and cosmology,
in use worldwide.
Following his appointment as
SDSC's chief scientific officer in June 2008,
Dr. Norman worked to foster collaborations
across the UCSD campus for
cyberinfrastructure-oriented
research,
development and education.
He also serves as division director of
SDSC's Cyberinfrastructure
Research, Education and Development (CI-RED),
and is the Principal Investigator for two of
SDSC's leading HPC systems
—
Gordon-Simons
and
Comet
—
which together represent
more than $42 million in
National
Science Foundation
funding.
Dr. Norman's work has earned him
numerous honors,
including Germany's prestigious
Alexander von Humboldt Research Prize,
the
IEEE
Sidney
Fernbach Award,
and several
HPCC Challenge Awards.
He also is a Fellow of the
American
Academy of Arts and Sciences,
and the
American
Physical Society.
He holds an M.S. and Ph.D. in
engineering and applied sciences
from
University
of California Davis,
and in 1984 completed
his post-doctoral work at the
Max
Planck Institute for Astrophysics
in Garching, Germany.
From 1986 to 2000,
Dr. Norman held numerous positions at
the
University
of Illinois at Urbana-Champaign,
as a
National
Center for Supercomputing Applications
(NCSA)
associate director and
senior research scientist under
Larry
Smarr,
currently UCSD's
director of the
California
Institute for Telecommunications and
Information Technology
(Calit2);
and as a professor of
astronomy.
From 1984 to 1986,
he was a staff member at
Los
Alamos National Laboratory.
Executive Director
Shodor
Education Foundation
Scientist in Residence
Wofford
College
Topic:
"What the Sciences Can Learn from Each Other:
An Exploration of Interdisciplinary
Supercomputing"
Slides:
Click
here.
Talk Abstract
Research and progress in
high performance computing
often require
multidisciplinary and interdisciplinary efforts
to take advantage of
model and simulation approaches.
In this talk,
I will explore some ideas
that demonstrate how
the thinking in one science
can influence a simulation in another.
We'll draw examples from biology and physics,
primarily,
to show what we can learn from each other.
Biography
Dr.
Robert M. Panoff
is founder and Executive Director of
the
Shodor
Education Foundation,
a non-profit education and research corporation
in Durham NC,
dedicated to reform and improvement of
mathematics and science education through
computational and communication technologies.
He serves as Undergraduate Education Director
for both
the
eXtreme
Science and Engineering Discovery
Environment
(XSEDE)
and the
Blue
Waters
Petascale
Computing
effort,
both in partnership with
the
National
Center for Supercomputing Applications
(NCSA)
at the
University
of Illinois at Urbana-Champaign
and funded by the
National
Science Foundation
(NSF).
He is currently Visiting Professor at
Wofford
College.
His work explores the interactions between
technology and education to develop
courses and interactive simulation modules
that combine
standards,
curriculum,
supercomputing
resources and desktop computers.
In recognition of Dr. Panoff's efforts
in college faculty enhancement and
curriculum development,
Shodor was named as an
NSF Foundation Partner
for the revitalization of
undergraduate education.
In 1998,
Shodor established the
Shodor Computational Science Institute,
which was expanded with NSF funding in 2001 to
become the
National
Computational Science Institute
(NCSI).
Shodor's
Computational
Science Education Reference Desk
(CSERD)
serves nearly 4 million webviews per month
as a
Pathway portal
of the
National
Science Digital Library.
Dr. Panoff consults at
several national laboratories
and is a frequent presenter at NSF workshops
on
visualization,
supercomputing,
and networking.
Dr. Panoff received his M.A. and Ph.D. in
theoretical
Physics
from
Washington
University in St. Louis,
with both pre- and postdoctoral work at the
Courant
Institute of Mathematical Sciences
at
New
York University.
In 2005,
Wofford
College,
awarded Dr. Panoff,
an honorary Doctor of Science degree,
in recognition of his leadership in
computational science education.
In 2014,
he was the recipient of the
SIGCSE
Outstanding
Contribution to Computer Science Education
Award.
Executive Director
Texas
Advanced Computing Center
The University
of Texas at Austin
Topic:
"The Next Generation NSF Computing Resource"
Slides:
available after the Symposium
Talk Abstract
Coming soon
Biography
Dan Stanzione is the Executive Director of the
Texas
Advanced Computing Center
(TACC)
at
The
University of Texas at Austin
and the Principal Investigator for
Wrangler.
He is also the PI for TACC's 10 PetaFlop
Stampede
supercomputer,
and has previously been involved in
the deployment and operation of the
Ranger
and
Lonestar
supercomputers at TACC.
He served as the Co-Director of
The
iPlant Collaborative,
an ambitious endeavor to build
cyberinfrastructure to address
the grand challenges of plant science.
Prior to joining TACC,
Dr. Stanzione was the founding director of the
Ira A. Fulton
High Performance Computing Institute
(HPCI)
at
Arizona
State University (ASU).
Before ASU,
he served as an AAAS Science Policy Fellow
in the
National
Science Foundation
and as a research professor at
Clemson
University,
his alma mater.
Assistant Vice President
–
Research Strategy Advisor
Information
Technology
Director
OU
Supercomputing Center for Education
& Research (OSCER)
Information
Technology
Associate Professor
College
of Engineering
Adjunct Associate Professor
School
of Computer Science
University
of Oklahoma
Joint Co-manager
(with
Dana
Brunson)
XSEDE
Campus Engagement
program
Topic:
"OSCER State of the Center Address"
Slides:
PowerPoint
PDF
Talk Abstract
The
OU
Supercomputing Center for
Education & Research
(OSCER)
celebrated its 17th anniversary
on August 31 2018.
In this report,
we examine
what OSCER is,
what OSCER does,
what OSCER has accomplished
in its 16 years,
and where OSCER is going.
Biography
Dr.
Henry Neeman
is the
Director of the
OU
Supercomputing Center for Education &
Research,
Assistant Vice President
Information Techology
–
Research Strategy Advisor,
Associate Professor in the
College
of Engineering
and
Adjunct Associate Professor in the
School
of Computer Science
at the
University of
Oklahoma.
He and
Dana
Brunson
have been appointed
joint co-leads of the
XSEDE
Campus Engagement
program,
which includes the
Campus
Champions.
He received his BS in computer science
and his BA in statistics
with a minor in mathematics
from the
State
University of New York at Buffalo
in 1987,
his MS in CS from the
University of
Illinois at Urbana-Champaign
in 1990
and his PhD in CS from UIUC in 1996.
Prior to coming to OU,
Dr. Neeman was a postdoctoral research
associate at the
National
Center for Supercomputing Applications
at UIUC,
and before that served as
a graduate research assistant
both at NCSA
and at the
Center for
Supercomputing Research &
Development.
In addition to his own teaching and research,
Dr. Neeman has collaborated with
dozens of research groups,
applying High Performance Computing techniques
in fields such as
numerical weather prediction,
bioinformatics and genomics,
data mining,
high energy physics,
astronomy,
nanotechnology,
petroleum reservoir management,
river basin modeling
and engineering optimization.
He serves as an ad hoc advisor
to student researchers
in many of these fields.
Dr. Neeman's research interests include
high performance computing,
scientific computing,
parallel and distributed computing
and
computer science education.
Professor
Department of
Computer Science
Kansas State
University
Director
Institute for Computational Research
Topic:
"Birds-of-a-Feather Session:
XSEDE Region 4 Campus Champions"
(with
BJ
Lougee)
BoF Slides:
available after the Symposium
Abstract
The
XSEDE
Campus
Champions
program supports campus representatives
as a local source of knowledge about
local, regional and national
High Performance Computing and
Cyberinfrastructure information,
including XSEDE resources.
We would like to invite everyone
—
any current Campus Champions,
possible Campus Champions,
and those who are just curious
—
to a
"Meeting of the Champions."
During our Birds-of-a-Feather session,
join your fellow
Region 4
Campus Champions
(from AR, KS, LA, MO, NE, OK and TX)
to discuss the program,
what you can expect to gain from participating,
what we hope to achieve
in the way of
both short term and long term goals,
future visions for the program,
etc.
Dr. Dan Andresen
(Kansas State
University)
and
BJ Lougee
(Federal
Reserve Bank of Kansas City)
will act as the emcees
for this meeting,
which is expected to draw attendance from
both current and prospective Campus Champions
from around the Region.
Response to these regional meetings
in other parts of the US
have been
very favorable,
so we're anxious to offer the opportunity
to all of you!
Biography
Daniel
Andresen, Ph.D.
is a professor of
Computing
& Information Sciences
at
Kansas
State University
and Director of the
Institute for Computational Research.
His research includes
embedded and distributed computing,
biomedical systems,
and high performance scientific computing.
Dr. Andresen coordinates the activities of
the K-State research computing cluster,
Beocat,
and advises the
local
chapter
of the
Association
for Computing Machinery
(ACM).
He is a
National
Science Foundation
CAREER
award winner,
and has been granted research funding from
the NSF,
the
Defense
Advanced Research Projects Agency
(DARPA),
and industry.
He is a member of
the
Association
for Computing Machinery,
the
IEEE
Computer Society,
the
Electronic
Frontier Foundation,
the
American
Society for Engineering Education,
and has been an
XSEDE
Campus
Champion
since 2011.
Undergraduate Student
Computer
Science Department
Midwestern
State University
Topic:
"Scientific Kernel Performance Evaluation
under CUDA and OpenACC: A Case Study"
(with
Eduardo
Colmenares)
Slides:
available after the Symposium
Abstract
This research studies
the behavior and performance of
two and widely used
computationally intensive kernels
in
two of the most popular
many-core programming models,
CUDA and OpenACC.
This research also shows that
the nature of the problem
plays a crucial role
in determining what many-core model
will provide the highest performance.
The first kernel under study is
a
Fast
Fourier Transform
(FFT),
which samples a signal over
a period of time
and
divides it into its frequency components,
computing the
Discrete
Discrete Fourier Transform
(DFT)
of a sequence.
The second kernel is
Matrix Multiplication,
which is a critical routine in
Mathematics,
Artificial Intelligence,
Machine Learning
and
Deep Learning.
Biography
Shady Boukhary
is a
Computer
Science
student at
Midwestern
State University.
He is very passionate about
technology and has always been fascinated by
its impact on daily lives.
His main interests are
High Performance Computing,
Deep Learning,
and
Software Engineering.
Shady Boukhary
has also shipped
multiple software solutions to
multiple software companies.
Director of Operations
and
Senior Research Scientist
Center for
Analysis & Prediction of Storms
University of
Oklahoma
Topic:
"Can a Fleet of Drones
Improve Your Weather Forecast?
Experiments with 3D Mesonet"
Slides:
PDF
Talk Abstract
For the past 25 years,
the State of Oklahoma has benefitted from
having a network of
120 surface weather observing stations,
known as the
Oklahoma
Mesonet,
deployed across the state.
It has been proposed that
a fleet of
automated Small Unmanned Aerial Systems (sUAS),
also known as drones,
could be
deployed at many,
or all,
of the Mesonet sites,
to make vertical profile
measurements of
temperature,
humidity
and
pressure
in the lower atmosphere,
thus creating a 3D Mesonet.
There are technical and policy impediments to
deploying such a network today,
but through the use of an
Observing System
Simulation Experiment,
or OSSE,
we can investigate the ability of such a
network
to improve the numerical weather forecasts
for critical severe weather events.
Here we present the results of a few tests
that were performed by
recently-graduated MS student
Andrew Moore,
utilizing
OSCER's Schooner supercomputer.
Biography
Keith Brewster is
Director of Operations
and
Senior Research Scientist
at the
Center
for Analysis and Prediction of Storms
at the
University
of Oklahoma
and an Adjunct Associate
Professor in the
OU School of
Meteorology.
His research involves
data assimilation of
advanced observing systems
for high resolution
numerical weather analysis and prediction,
including data from
Doppler
radars,
satellites,
wind profilers,
aircraft
and
surface mesonet systems.
He earned an M.S. and Ph.D. in Meteorology
from the
University
of Oklahoma
and a B.S. from the
University
of Utah.
Assistant Professor
Computer
Science Department
Midwestern
State University
Topic:
"Scientific Kernel Performance Evaluation
under CUDA and OpenACC: A Case Study"
(with
Shady
Boukhary)
Slides:
available after the Symposium
Abstract
This research studies
the behavior and performance of
two and widely used
computationally intensive kernels
in
two of the most popular
many-core programming models,
CUDA and OpenACC.
This research also shows that
the nature of the problem
plays a crucial role
in determining what many-core model
will provide the highest performance.
The first kernel under study is
a
Fast
Fourier Transform
(FFT),
which samples a signal over
a period of time
and
divides it into its frequency components,
computing the
Discrete
Discrete Fourier Transform
(DFT)
of a sequence.
The second kernel is
Matrix Multiplication,
which is a critical routine in
Mathematics,
Artificial Intelligence,
Machine Learning
and
Deep Learning.
Biography
Dr. Eduardo Colmenares is
an Assistant Professor of
Computer
Science
at
Midwestern
State University.
He received his BS in
Electronics Engineering
from the
Industrial
University of Santander,
Colombia,
his Master of Science and PhD in
Computer
Science
from
Texas
Tech University,
both with Focus in
High Performance Computing and
Scientific Computing.
Dr. Colmenares serves as
a member of the steering committee for
the
Consortium
for Computing Sciences in Colleges
(CCSC),
and is also the author of
multiple peer reviewed publications
in HPC and Software Engineering.
His research interest include
HPC,
Deep Learning
and
undergraduate Software Engineering.
IT Manager
Information Technology
Laureate
Institute for Brain Research
Topic:
"Making an HPC Storage Procurement Decision
Informed by HPC"
Slides:
available after the Symposium
Abstract
Laureate Institute for Brain Research
faces yet another storage challenge in 2018.
This challenge is created by
the existence of
a sizable amount (160 TB) of
scientific data
made up of files
ranging in size between
20 KB and 80 KB
that need to be available for analysis.
While this data can be stored in
our 5 PB nearline storage environment
made up of HDDs and Ceph,
performance for analysis
is not what our users would prefer.
Furthermore,
LIBR operates
a VMware environment
consisting of approximately 150 guests
consuming 50 TB of storage.
Both scenarios would benefit greatly from
a flash storage offering.
Unfortunately,
200+ TB of flash storage is not a cheap ask.
As such,
LIBR embarked on
an effort to add flash storage to
its existing Ceph storage environment.
Because Ceph generally is not
a commercial product with
standardized hardware
and
known performance characteristics,
LIBR must determine
how its cluster will perform
with specific disks under specific workloads.
This talk covers
how LIBR is leveraging its HPC environment
to analyze
over 20 billion benchmark data-points
created with a battery of tests
against each candidate model of SSD
being evaluated in this project.
Biography
Brady
Deetz
is the head of IT for the
Laureate Institute for Brain Research,
a clinical neuroscience research institute in
Tulsa OK.
At LIBR,
he is responsible for
research computing and data storage
in support of
multidisciplinary research programs in
neuroimaging,
behavioral health,
and
biochemical measures.
Key cyberinfrastructure resources include
petascale tape storage;
a petascale disk (Ceph) resource,
and local compute servers.
System Administrator
Department of
Computer Science
Kansas State
University
Topic:
"Birds-of-a-Feather Session:
HPC System Administrators"
Slides:
none
Abstract
We recently observed that
system administrators
are often leary of advice
(even from their direct supervisors)
unless it has been given from
somebody else who
(a) has signficant relevant experience,
(b) they know and trust,
and
(c) can point out the "gotchas."
This Birds-of-a-Feather session
is for system administrators
to talk among themselves about
current pain-points,
things we've done that have turned out well,
and
things we've done that have flopped.
Biography
Kyle Hutson has been involved with
Linux system administration since 1994.
He received his bachelor's degree from
Kansas State
University
in
computer
engineering
in 1995.
He has worked in
non-profit,
public sector,
and
private sector IT services,
including several years as
a small business IT consultant.
Kyle joined
Kansas State University's
HPC team in 2012.
Director, Global Sales Strategy for
High Performance Computing &
Artificial Intelligence
ISG
Dell
EMC
Topic:
"Can HPC technologies lead the way for AI?"
Slides:
PDF
Talk Abstract:
Deep Learning (DL)
is the new posterchild for
cutting-edge technologies,
with AI-fueled breakthroughs
being announced almost on a daily basis.
Many of the algorithms behind DL
are several decades old,
and advances in
affordable hardware technologies,
coupled with copious amounts of usable data,
has led to the recent resurgence in interest.
Even though we're still in the early days of
DL as a technology,
researchers are already running into
hurdles of scalability.
For the practitioners in HPC,
this is nothing new,
where the argument of scale-up vs scale-out
has played out
has seen a clear winner
dictated more by practicality
rather than technology.
Is Deep Learning at a similar juncture,
and what can our experience with HPC
tell us about where it's headed?
Biography
Adnan Khaleel
is a seasoned business development professional
with over 18 years of experience in
High Performance Computing
and
Machine Learning.
Currently,
Adnan is leading Global HPC Sales Strategy at
Dell EMC,
focusing on Sales Strategy for HPC & AI.
Due of his technical background,
Adnan often works very closely with customers
in helping them choose
optimal HPC solutions and architectures.
Prior to Dell EMC,
Adnan was the product manager at
Cray
responsible for
big data analytics,
and led the development and launch of
the award winning
Urika-XA™
High Performance Data Analytics
(HPDA)
platform.
Prior to Cray,
Adnan served as the chief technical marketer at
Newisys,
an Austin TX server startup,
and prior to that,
as a member of the Hardware Architecture groups
at both
HaL
computers and
Motorola,
where he developed
innovative simulation tools for
architectural studies on
next-generation massively parallel systems.
Adnan has
a deeply technical engineering background
in designing
SuperScalar processors,
high-speed computer networks,
and
massively scalable systems design.
He has also published several papers on
computer networks,
and holds a U.S. patent on
multiprocessor technologies.
He has an MBA from the
McCombs
School of Business
at
University
of Texas
and an MS in
Electrical
Engineering
from
Texas
A&M University.
Research Data Specialist
University
Libraries
University
of Oklahoma
Topic:
"What's New with the Carpentries at OU:
A Status Update"
Slides:
available after the Symposium
Abstract
For nearly five years,
with the last three as a foundation member,
the University of Oklahoma
has been associated with
Software
Carpentry,
and recently the
Carpentries
Foundation.
In January 2018,
Software Carpentry
and
Data
Carpentry
merged to form a new
Carpentries Foundation.
This breakout session will start with
a report on the merger,
and the current state of the
OU
Carpentries
over the past three years of membership.
An update will be given on
the current status of
the renewal process with
the Carpentries Foundation.
There are two new domains
starting the process to join
Software and Data Carpentries
as part of the
Carpentries Foundation.
Library
Carpentry
and
HPC
Carpentry
groups have been meeting,
developing lessons and curriculum,
following the Carpentries
pedagogy and lesson development guidelines.
Library Carpentry
is further along in the process,
as governance elements are being formed
to help guide the community.
A brief outline of future activities of the
OU Carpentries
and how one can get involved,
and the announcement of new domain lessons
(geospatial for Data Carpentry)
as well as updates to existing lessons
will finish the session.
Biography
Dr. Mark Laufersweiler
has always had a strong interest in
computers,
computing,
data
and
data visualization.
Upon completing his post-doc
work for
the
Atmospheric
Radiation Measurement
(ARM)
program,
he was the lead computer systems administrator
for 3.5 years
serving the
Florida
State University
Department
of Earth, Ocean and Atmospheric Science.
He was then
the Computer Systems Coordinator for
the
University
of Oklahoma
School
of Meteorology
from 1999-2013.
Part of his duties included
managing the real time data feed
and maintaining the departmental data archive.
He assisted with faculty
in their courses
to help foster computing skills
needed for the classroom
and instruction based on
current best practices regarding
research data and code development.
Since the Fall of 2013,
he has served as the
Research Data Specialist
for the
University
of Oklahoma Libraries.
He is currently assisting
the educational mission of the Libraries
by developing and offering
workshops,
seminars
and
short
courses,
helping to inform
the university community
on best practices
for
data management and data management planning.
He is also working on
the formation of a data repository
to host research data generated by
the university community.
He is a strong advocate of
open source software
and
open access to data.
In 2008,
Dr. Laufersweiler was
awarded
the
Russell
L. DeSouza Award.
This award,
sponsored by
Unidata
is for individuals whose
energy,
expertise,
and active involvement
enable the Unidata program
to better serve geoscience.
Honorees personify Unidata's ideal of
a community that shares
data,
software,
and
ideas
through computing and networking technologies.
Cyberinfrastructure Engineer and
Cyberinfrastructure Practitioner
Center
for the Advancement of Data and Research
in Economics
(CADRE)
Federal
Reserve Bank of Kansas City
Topic:
"Birds-of-a-Feather Session:
XSEDE Region 4 Campus Champions"
(with
Dan
Andresen)
BoF Slides:
available after the Symposium
Abstract
The
XSEDE
Campus
Champions
program supports campus representatives
as a local source of knowledge about
local, regional and national
High Performance Computing and
Cyberinfrastructure information,
including XSEDE resources.
We would like to invite everyone
—
any current Campus Champions,
possible Campus Champions,
and those who are just curious
—
to a
"Meeting of the Champions."
During our Birds-of-a-Feather session,
join your fellow
Region 4
Campus Champions
(from AR, KS, LA, MO, NE, OK and TX)
to discuss the program,
what you can expect to gain from participating,
what we hope to achieve
in the way of
both short term and long term goals,
future visions for the program,
etc.
Dr. Dan Andresen
(Kansas State
University)
and
BJ Lougee
(Federal
Reserve Bank of Kansas City)
will act as the emcees
for this meeting,
which is expected to draw attendance from
both current and prospective Campus Champions
from around the Region.
Response to these regional meetings
in other parts of the US
have been
very favorable,
so we're anxious to offer the opportunity
to all of you!
Biography
BJ Lougee
is a computer scientist and
cyberinfrastructure engineer and practitioner
in the
Center
for the Advancement of Data and Research
in Economics
(CADRE)
at the
Federal
Reserve Bank of Kansas City.
He is also the
XSEDE
Campus
Champion
Deputy
Director for Region 4.
Prior to joining the
Bank in 2014,
he was the Lead HPC Systems Administrator at
the
High Performance Computing Center (HPCC)
for
the
76th Software Maintenance Group
at
Tinker
Air Force Base.
He holds a Bachelor of Science in
Computer
Science
from the
University
of Central Oklahoma.
He is currently working
toward his Master of Science in
Computer
Science,
with an emphasis in
HPC and machine learning,
at
Georgia
Institute of Technology.
He conducts research and development on
the Bank's
HPC environment,
and trains researchers on using HPC.
He has a particular research interest in
helping to drive the adoption of
HPC techniques in the economics field.
Associate Director for
Research Computing Strategy
OU
Supercomputing Center for
Education & Research
(OSCER)
University
of Oklahoma
Topic:
"Industrial HPC for Fun and Profit"
Slides:
available after the Symposium
Talk Abstract
Telling stories about
technical computing's application to
common products and services
is an effective way to provide
relatable touchstones
when communicating about HPC.
This talk presents eight easily understood,
and hopefully entertaining,
stories about
how high performance computing
contributes to people's daily lives,
with examples ranging from
cars and trucks to
laundry detergent and coffee cans,
to low flow toilets.
The purpose is to equip attendees with
examples of useful stories and
examples that they can employ in
their own outreach and education efforts.
Biography
George Louthan
is the
Associate Director for
Research Computing Strategy
of the
OU
Supercomputing Center for
Education & Research
(OSCER)
at the
University
of Oklahoma.
Before coming to OU,
he served as the Director of the
Tandy
Supercomputing Center,
an initiative of the
Oklahoma
Innovation Institute
in Tulsa OK.
He holds a MS in Computer Science and
undergraduate degrees in
Computer Science and Mathematics from the
University
of Tulsa.
Before moving to high performance computing,
his background included work in
information security,
research software development
and
informatics.
Associate Professor
School
of Computer Science
Department
of Microbiology & Plant Biology
University
of Oklahoma
Topic:
"Disco: A Distributed High-Performance
Assembler for Large Metagenomes"
Slides:
available after the Symposium
Talk Abstract
A distributed,
iterative,
scalable co-assembler,
Disco,
was developed to assemble
large real-world metagenomes
from up to billions of
Illumina
reads.
Disco was parallelized using
a hybrid architecture that integrated
shared-memory multi-threading,
point-to-point message passing,
and remote memory access.
Metagenome assembly and scaffolding
were performed
using an iterative overlap graph approach.
Disco was compared with
metaSPAdes,
ABySS,
MEGAHIT,
MetaVelvet,
and
Ray Meta,
using three mock communities
spanning a range of complexities.
The assembly quality of Disco
was comparable to
the best of these assemblers,
based on
completeness,
accuracy,
and contiguity of their assemblies
of these mock communities.
The computational performance of Disco
was evaluated on hardware
from single computers
to high-performance-computing clusters.
On single computers,
Disco was more computationally efficient than
metaSPAdes,
ABySS,
MEGAHIT,
and
MetaVelvet.
On computer clusters,
Disco also scaled better than
Ray Meta,
ABySS
and
HipMer.
These advantages of Disco
enabled high-quality assembly of
real-world metagenomes
from deep sequencing of
complex communities in
wetland sediments and human gut.
Disco is easy to use and freely available
here.
Biography
Dr. Pan recently joined the
School
of Computer Science
and the
Department
of Microbiology & Plant Biology
at the
University
of Oklahoma.
he was a senior research scientist in the
Computer
Science and Mathematics Division
at
Oak
Ridge National Laboratory.
The research in Dr. Pan's lab
is focused on using
high-performance computing,
machine learning,
and network analysis
to analyze big -omics data
from microbial communities
and eukaryotic organisms.
He has developed scalable algorithms
to assemble and annotate large metagenomes
and
process mass spectrometry data
from proteomics and metabolomics.
These informatics tools
have enabled many biological discoveries
regarding nutrient cycling in
soil and metabolic interactions
between plants and
their rhizosphere communities.
C. M. Sliepcevich Professor of
Chemical Engineering
School
of Chemical, Biological &
Materials Engineering
University
of Oklahoma
Topic:
"Coarse-Grained Computations for
Understanding Nanofluids"
Slides:
available after the Symposium
Talk Abstract
Coarse-grained computations using
Dissipative
Particle Dynamics
(DPD)
techniques
can provide insights to
complex physical processes,
where molecular level simulations
are too cumbersome
and
macroscopic simulations can miss
the nuances of the physics occurring at
the molecular level.
We use DPD techniques
to simulate nanofluidic behavior
on OSCER machines.
In this talk,
we will present
a protocol for the determination of
model parameters for DPD computations,
and recent results about
the behavior of carbon nanotubes
with surfactants in aqueous suspensions
and
at the oil-water interface.
This is important in several application areas,
including
oil recovery
and
industrial separation processes.
Biography
Dimitrios Papavassiliou
is the
C. M. Sliepcevich Professor in the
School
of
Chemical, Biological & Materials
Engineering
at the
University
of Oklahoma.
He received a BS degree from the
Aristotle
University of Thessaloniki,
and MS and PhD degrees from the
University
of Illinois at Urbana-Champaign.
Engineering at the University of Oklahoma.
Dimitrios joined the University of Oklahoma
after working at
Mobil's
Upstream
Strategic Research Center
in Dallas, Texas.
His research contributions
are in the area of
computations and numerical methods
for turbulent flows and flows in porous media,
in the area of micro- and nano-fluidics,
and in the area of biologically relevant flows.
He is the author or co-author of
over 115 archival journal publications.
His research work has been funded by
federal institutions,
private consortia of companies
and
private foundations.
Dr. Papavassiliou has served as the
Program Director for
at the
National
Science Foundation
between 2013 and 2016.
He is currently a member of the
Consulting Editors Board
of the
AIChE
Journal,
a member of the
Editorial Board
of the journal
Fluids,
and is a Fellow of the
American
Institute of Chemical Engineers.
Senior Fellow & Chief Technology Officer
Formulus
Black
Talk Topic:
"Persistent Memory & HPC -
the Next Frontier"
Talk Slides:
available after the Symposium
Abstract
This talk will deliver
a brief overview of the current state of
the art for persistent memory technologies,
followed by a discussion of
breaking the memory bottleneck
currently experienced by
many HPC architectures,
including both
CPU and GPU-based
compute techniques.
Biography
Robert Peglar
is currently the President of
Advanced
Computation and Storage LLC,
after a distinguished 40-year
career serving
the compute and storage technology industries,
spanning six companies.
His most recent post was
Senior Vice President and
Chief Technology Officer of
Formulus
Black,
leading development efforts in
next-generation software
for persistent in-memory computing.
Before that,
he was Vice President of Advanced Storage at
Micron
Technology,
where he led efforts in
advanced storage systems strategy,
led the CTO function and
executive-level planning with
key customers and partners worldwide for
Micron's Storage Business Unit,
and defined future
storage portfolio offerings.
Prior to joining Micron in February 2015,
Mr. Peglar was
Chief Technology Officer, Americas at
EMC Isilon
for four years.
Prior to EMC,
he was a Senior Fellow and
Vice
President of Technology at
Xiotech
for 11 years.
Prior to Xiotech,
he held key
technology specialist and
engineering leadership positions
over a 10-year period at
StorageTek
and its subsidiary,
Network Systems Corporation.
Prior to StorageTek,
he held
engineering development and product management
positions for a decade at
Control Data
and its supercomputer division,
ETA Systems.
Mr. Peglar serves on the Board of Directors of
the
Storage
Networking Industry Association
(SNIA),
is the former co-chair of the
SNIA Analytics and Big Data Committee
and the
SNIA
Tutorials,
and is the former
director of the
SNIA
Solid State Storage Initiative.
He also serves as an advisor to the
Flash
Memory Summit.
Mr. Peglar has extensive experience in
persistent memory architecture,
data management and analysis,
high-performance computing,
nonvolatile systems,
distributed cluster architectures,
filesystems,
I/O performance
optimization,
cloud storage,
replication and archiving strategy,
networking protocols,
and storage virtualization.
He is
a sought-after keynote speaker and panelist at
leading storage and computing-related seminars
and conferences worldwide.
Mr. Peglar is a four-time EMC Elect honoree,
2014-2017.
He was also one of
25 senior executives worldwide
selected for the
CRN
2010 Storage Superstars Award.
Mr. Peglar
holds a B.S. degree in
Computer
Science
from
Washington
University in St. Louis,
and performed graduate work at
Washington University's
Sever
Institute of Engineering.
His research background includes
memory optimization,
distributed systems,
I/O performance analysis,
queuing theory,
parallel systems
architecture and operating system design,
filesystems and storage networking protocols,
clustering
algorithms and virtual systems communication.
OTHER
BREAKOUT SPEAKERS TO BE ANNOUNCED