xt74j09w4064 https://exploreuk.uky.edu/dips/xt74j09w4064/data/mets.xml University of Kentucky. Office of the Vice President for Research and Graduate Studies Kentucky University of Kentucky. Office of the Vice President for Research and Graduate Studies 2002 volumes. Call Number: LD2762.9 .K38c. Accession Number: 2018ua049 journals LD2762.9 .K38c English Office of the Vice President for Research and Graduate Studies  Contact the Special Collections Research Center for information regarding rights and use of this collection University of Kentucky Odyssey Magazine Research -- Kentucky -- Periodicals Education -- Research -- Periodicals Research -- Kentucky Odyssey: The Magazine of University of Kentucky Research and Graduate Studies, Spring 2002 text images Odyssey: The Magazine of University of Kentucky Research and Graduate Studies, Spring 2002 2002 2002 2026 true xt74j09w4064 section xt74j09w4064  

 

  

UK

Lee T. Todd jr.. President

Jim Boling, Acting Vice President for Research
Joseph L. Fink 111. Associate Vice President for Research
Delwood C. Collins, Associate Vice President for Research‘Medical Center

Agriculture

M. Scott Smith. Dean
Allied Health

Thomas C. Robinson, Dean
Architecture

David Mohnev. Dean

Arts and Sciences
Howard Grotch. Dean

Business and Economics
Richard W. Furst, Dean

Communications and Information

Studies
j. Davrdjohnson. Dean

Dentistry
Leon A. Assael. Dean

Education

James Cihulka. Dean
Engineering

Thomas \X. Lester. Dean
Fine Arts

Robert Shav. Dean
Human Environmental Sciences
Retia Walker. Dean

law

Allan W. Vestal Dean
Medicine

Emery A. \V'ilson. Dean
Nursing

Carolyn A. Williams. Dean
Pharmacy

Kenneth 15 Roberts, Dean

Social Work
Kay Hoffman. Dean

 

The warld’sfirst
medical monitor
for heart failure,
the Chronicle IHM
is allowing patients

to lead normal lives.
UK is one of only eight
centers across the
country approved to
implant the device.

’ .a ~ ODYSSEY

 

Director

Editor

Associate Editor
and Web Designer

Designer
Photographer
Printer

The Magazine of University Of Kentucky Research

ODYSSEY

SPRING 2002

IN THIS ISSUE

 

FOR PATIENTS WITH NEW HEART
DEVICE THE BEAT GOES ON 2

WORKING TO SOLVE
THE FOAL DEATHS MYSTERY 6

SUPERCOMPUTERS “R” US:
LINKING PCS FOR
UNPRECEDENTED POWER 8

FIRED—UP FOR THEIR ART 14

WORKING TO BREAK
INDIA’S LEGAL LOGJAM 18

PATTERSON SCHOOL HELPING
KENTUCKY PREPARE FOR THE WORST 21

  

Published twice a year b3 the Office of the Vice President for Research. ODYSSEY seeks to
inform, educate and entertain you on your journey through some of the cutting—edge research
that is part of the l‘niversitv of Kentucky's SZ<)Oimillionatjvear research enterprise

To receive a free Subscription. contact tKleSev’dtikvetlu or [K Research Communications
ODYSSEY. ZZZ Bowman llall. lfnhersrtv of Kentucky. Lexington. KY 4030000”. 83‘) 23",
829". ODYSSEY is also available on the Web at \th \\ l‘gSlllx'Y edu

Deborah \X’eis
jefl' Worley
Alicia P. Gregory

Shear 8r Shear
Lee R Thomas Photography
Gateway Press

@2002 University of Kentucky

  

 
 
  
 
  
 
  
 
 
  
 
 
   

NOTEWORTHY PRIMA DONNA
ENJOYING NEW ROLE AT UK

NEW RECRUITS ADVANCE
PLANT SCIENCE

2001—2002
UNIVERSITY RESEARCH PROFESSORS

BOBBIE ANN MASON:
ZIGZAGGING HER WAY
BACK TO UK

THE WELL-READ LIFE:
UNIVERSITY PRESS DIRECTOR
KEN CHERRY RETIRES

ODYSSEY TURNS 20

  

 

usan Levy, a 62-year—old mother of five

and grandmother of two, rang in 2002

in Vail, Colorado, celebratingwith fam—
ily and friends. Later that month, she flew
to California to see her brother. When she
is not traveling, Levy is likely to be found
showing homes to a real—estate client, at the
bridge table with friends, or at temple with
her granddaughter. She has standing get—
togethers with other groups of friends on
Tuesdays, Wednesdays, and Fridays. Her
life, she happily admits, is teeming with
activity.

Just a year ago, however, life was dra—
matically different for Levy. She couldn’t
walk across a room without having diffi—
culty breathing. She was tied to an oxygen
tank and slept on a hospital bed at home to
keep her head elevated enough so she could
breathe. In short, she was not able to lead a
normal life.

Levy had been diagnosed with heart fail—
ure, a diagnosis she shared with almost five
million Americans, including an estimated
60,000 Kentuckians. In fact, over 400,000
new cases of heart failure are diagnosed
each year, a 160 percent increase over the
past two decades. This increase is due,
ironically in part, to improved treatment of
heart attacks, although longer life spans
and diets high in fat also play a significant
role in this increase.

Despite its name, heart failure is not a
sudden stop in the heart’s ability to keep
beating. Rather, it is the heart’s inability to
pump powerfully enough to circulate oxy-
gen—rich blood adequately throughout the
body. Symptoms include shortness of
breath, fatigue, and fluid accumulation.

Heart failure is the most common cause

For Patients with
New Heart Device the

Beat Goes On

WITH THE
CHRONICLE
IHM, THE
WORLD’S FIRST
MEDICAL
MONITOR FOR
PATIENTS WITH
HEART FAILURE,

 

DOCTORS CAN MEASURE
THEIR PATIENT’S CONDITION
FROM ANYWHERE BY USING A
SECURE WEB SITE.

   

 

 

-_ 4-,. ,W- “.4

of hospitalization for adults in the United
States and is the only type of heart disease on
the rise. The estimated cost ofinpatient and
outpatient care is $56 billion a year. In Ken—
tucky, heart disease is the leading cause of
death among men and women, black and
white, and in the United States about 250,000
people die of heart failure each year.

And for those who live, the impact ofheart
failure on the quality oflife can be grim. The
disease typically impairs physical functions
and can lead to anxiety and depression.

Levy knows these facts all too well, but
today these realities come to mind only when
she has an occasional setback in her otherwise
good condition. “I’m a healthy person,” she
says. “I don’t even think of myself as sick.”

The marked difference came from a new

 

investigational device the size of a typical
pacemaker known as the Chronicle Implant—
able Hemodynamic Monitor (Chronicle IHM,
Medtronic Inc). It is the world‘s first im—
planted monitor for the medical management
of patients with heart failure.

Levy was one ofthe first two Kentuckians to
have the device implanted in her chest and one
of approximately 100 Chronicle IHM pa—
tients nationwide. The Linda and Jack Gill
Heart Institute at the University of Kentucky
is one ofonly eight centers across the country
approved to implant the device, and the only
site in Kentucky.

With the Chronicle IHM, physicians can
monitor their patient’s condition from any—
where by using the World Wide Web. The

patient simply passes a magnetic device over

 

2 ODYSSEY

 

 

  

 

DENNIS MCDONALD (FAR RIGHT) SAYS HE “GOT A LOT OF GOOD USE" OUT OF

THE CHRONICLE IHM DEVICE BEFORE HIS RECENT HEART TRANSPLANT. HE IS

PICTURED HERE AFTER HIS DAUGHTER AMY'S WEDDING. LEFT TO RIGHT ARE

GRANDDAUGHTERS MEGHAN AND EMILY. HIS WIFE AGNES. AND HIS GRAND-

SONS JESSE AND TEDDY.

 

SUSAN LEVY (LEFT) WAS ONE
OF THE FIRST TWO KENTUCK-

IANSTO HAVE THE CHRONICLE
IHM IMPLANTEDINHERCHEST.
SHE IS PICTURED HERE WITH
HER DAUGHTER. CATHY
WOLFSON. AND HER GRAND-
son. ELI. AT Slx FLAGS GREAT
AMERICA NEAR CHICAGOLAST
OCTOBER. SIX WEEKS AFTER
HAVING THE IMPLANT.

the implant. and electronic signals from the
Chronicle IHM are transmitted over telephone
lines to a Web site accessibie only to the physi—
cian and his staff. The physician receives vital
information about the patients condition. in-
cluding heart rate. body temperature. fluid
levels. pulse. and. most importantly, the dias—
tolic and systolic pressures in the heart—im—
portant measures of the heart‘s strength.

William T. Abraham. co—director ofthe Gill
Heart Institute and chief of cardiovascular
medicine, is the principal investigator for the
study at UK and a member of the national
committee overseeing development ofthe de-
vice. On August 22 oflast year. Westby Fisher.
a UK electrophysiologist. implanted the device
in Levy and Wanda Moore. By the end of
December. UK physicians had implanted the
lHM in 15 more patients.

“This device has the potential to revolution—
ize the way we take care of patients." says
Abraham. “The biggest advantage is that we
can monitor remotely what is going on with the
patient‘s heart. We can ltnow how they are
doing in another part of the state without
having them go to an emergency room or drive
to Lexington."

(on/{Hutw/ on page a

 

 

More Heart Devices
in Clinical Trials

The Chronicle IHM is not the only heart
device undergoing clinical trials at the Uni-
versity of Kentucky Hospital. William T.
Abraham says UK physicians are testing
around 20 diagnostic or therapeutic devices
and investigational drugs—all aimed at help—
ing patients with heart failure.

Abraham led the U.S./Canadian trial of
an implantable device known as Insync, and
the FDA has now approved that device for
general use. Similar in size and weight to a
pacemaker, this device resynchronizes the
beating of the patient's right and left ven—
tricles—the heart’s pumping chambers. This
correction increases pumping power and
improves cardiac output, thereby allowing
blood to circulate throughout the body.

Abraham designed and organized the clini—
cal trial for Insync, taking the device from
concept to F DA approval. The clinical trial
followed severely ill patients for almost two
years. At the end of six months. 68 percent
of patients were judged to be improved. and
33 percent who used the device were able to
walk significantly longer distances.

“For many patients, that is the difference
between performing almost no daily activi—
ties and performing some or all daily activi-
ties,” Abraham said during a presentation at
a medical conference last March.

Insync 1CD, a second-generation Insync
device, is also now in clinical trial at UK.
Insync 1CD not only resynchronizes the
heart, but also incorporates defibrillator tech—
nology for patients prone to sudden cardiac
arrest.

“Heart failure is a serious problem in
Kentucky,” notes Abraham. “Kentucky
ranks first, second, or third in every measure
when it comes to the prevalence of heart
disease. These devices make it possible to
treat more heart—failure patients. Our goal is
for UK to become a prominent center for
the treatment of heart failure."

 

 

 

i'NIVERSI’I‘Y ()F KENTI'CKY

3

 

  

 

 

Quicker Treatment for Heart Attack Victims

How do you know if you are having a heart attack? Do you
know what to do when you are having symptoms?

Many people don’t know and even those who do often delay
seeking treatment, wasting valuable time that could save heart
muscle or perhaps their life, according to Debra K. Moser,
professor and Linda C. Gill chair of Cardiovascular Nursing in
the College of Nursing.

Moser says that each year about one million Americans have
an acute myocardial infarction, commonly known as a heart
attack. Up to 45 percent of these people will die, 60 percent of
them before they ever get to a hospital. But if they sought
immediate treatment for their symptoms and received either
thrombolytic therapy (also known as “clot—busting”), which uses
an intravenous drug to dissolve the clot in the coronary artery
that is stopping blood flow to the heart, or other therapies
(includingangioplastyorcoro—
nary artery bypass), the sever—

treatment, ideally reducing it to under an hour.

In this program, nurses are following high—risk patients—
typically those diagnosed with coronary artery disease, peripheral
arterial disease, or clinical atherosclerotic disease—for 18 months.
An advanced practice nurse meets with the patient in his or her
home within a couple days after leaving the hospital. (This is the
time when patients are most anxious and most likely to have
problems following their physician’s orders.) Nurses continue to
monitor the patient with telephone calls and with further home
visits if needed.

While there is certain information all patients receive, the
meetings are individualized to meet each patient’s needs. On
average, patients have an hour of in—person intensive instruction
followed by short phone calls, including discussions about typical
symptoms, sessions in problem solving, and information on the

potential benefits of acting quickly.
“By meeting individually with

 

ity of the heart attack could be
reduced, preventing irrevers—
ible damage and dysfunction.

Time is the enemy for some-
one who has had a heart at-
tack. “The best outcomes are
seen in patients who receive
treatment within one to two
hours of symptom onset,” says
Moser. “The problem is, most
people wait between two to
four hours before seeking treat—
ment.” In fact, only five per—

 

patients, we can respond to their
concerns and questions,” says
Moser. “It is far more interactive
and allows us to reinforce the in—
formation they receive in the hos—
pital.”

The goal for the study is two—
fold, Moser says. Through this
clinical trial, she hopes to decrease
the delay time for patients seeking
treatment and to increase use of
emergency medical systems such
as ambulances.

 

cent of eligible patients receive
thrombolytics within an hour

DEBRA MOSER SAYS THAT IF HEART ATTACK VICTIMS
SEEK IMMEDIATE TREATMENT, THE SEVERITY OF THE
ATTACK CAN BE REDUCED, PREVENTING IRREVERSIBLE

“The biggest part of the delay in

getting appropriate treatment is

 

and less than 25 percent re-
ceive any thrombolytic therapy
at all. One-fourth to one-half
of patients delay four hours before seeking treatment.

“When people experience pain, they typically try to think it
away or wish it away,” says Moser. “They try to relax or change
positions. Some just pretend nothing is wrong, and others
attribute it to more benign causes. Some try home remedies or
take a Tylenol. When they do decide to seek treatment, many
will drive themselves to the hospital.”

Public information campaigns, including television commer—
cials and print advertising, have had little impact on this prob—
lem, so Moser is hoping that their new program involving
education and counseling intervention delivered by nurses will
reduce the amount of time between the onset of symptoms and

DAMAGE.

not waiting for an ambulance or
waiting to be treated once at the
hospital," Moser notes. “The most
devastating delay comes from the patient not making a quick
decision to seek treatment. If we can reduce this decision time, we
can improve outcomes.”

The UK program is part of a $2.8 million National Institutes
of Health five-year study to test the effectiveness of health—
administered education at five sites around the world.

   

 

 

UK’s Reducing Treatment Delayfiir Heart Altar/e
Symptoms program is accepting patients into the
trial lfyou laaue been diagnosed wit/7 heart disease
and would like to participate, call toll flee 866-
570—8528. There is no cost to participate in the study.

 

 

 

ODYSSEY

 

 

  

 

 

 

 

WILLIAM T. ABRAHAM. THE GILL FOUNDATION PROFESSOR
IN INTERVENTIONAL CARDIOLOGY

The Chronicle [HM also greatly improves the physician‘s diag—
nostic capabilities. Making treatment decisions about heart patients
is often difficult. Abraham says. ”We may think patients are retain—
ing too much fluid when they are really dehydrated. W’ith this
device, we know exactly what is going on with fluid levels and other
critical measures."

Another patient and a pepperoni pizza gave the Chronicle IHM
a good test.

"He called in not feeling well, and thanks to the device, we knew
he had greatly elevated fluid levels," Abraham recalls. ”We kept
quizzing him about what he had done or eaten, but he denied doing
anything out ofthe ordinary. We knew his numbers were not right.
told him what medication to take. and his condition improved. Later
he confessed that he had been anxious so he had eaten an entire
pepperoni pizza. All that salt caused him to retain too much fluid."

Although she has no plans to eat an entire pizza in one sitting any
time soon, Levy says she has had similar concerns.

“My symptoms are not always the same as other people’s," she
says. “Sometimes I appear to be dehydrated when I am actually

 

“This device has the potential

to revolutionize the way we
take care of patients. ”

—\\'illiam T. Abraham

overloaded on fluids and need to get rid ofwater. This device tells
my doctors what is happening. I haven't had to go to the hospital
since I‘ve had it. I dona think I would have survived without it.“

Levy was indeed disabled by heart failure, according to Abraham,
but she was also a good candidate for the device because she has what
cardiologists call moderate-to—severe heart failure. If her case had
been mild, she could have probably been treated with medication
alone. Ifshe had been in end—stage heart failure, her probability for
success with the device wouldn‘t have been good either, and she
would have likely faced multiple hospitalizations.

Keeping patients out of the hospital will likely nullify the esti—
mated SI0.000—SIZ,000 the device will cost ifand when it receives
FDA approval. Abraham says that this cost would be about the
equivalent of two typical hospital stays for a patient with heart
failure.

Patients do not even need to be admitted to have the Chronicle
IHM implanted. The one—hour surgery is an outpatient procedure
during which the oval—shaped device is implanted just below the
collarbone and a single wire is run to the heart. There is less than a
one percent risk ofinfection from the implantation and minor risks
of local bleeding, the only complications found thus far. N0 long—
term complications have been discovered during the Phase II clinical
trial: Abraham expects to begin Phase III trials this spring.

Perhaps the greatest benefit of the device is its potential to let
patients lead normal lives again.

”They push me to have quality oflife," Levy says ofAbraham and
her other physicians and nurses. “I‘m basically having a wonderful
life. IfI have a problem, I call them, and they tell me what to do. The
Chronicle has given me my life back. I can do anything I want to
now. It has been a miracle."

 

I'XIVERSITY OF KENTI‘CKY 5

 

 PHOTO BY STEVE PATTON

 

  

ALTHOUGH THE CATERPILLAR-
CYANIDE THEORY HASN’T HELD
UP IN RECENT LAB TESTS, THE
EASTERN TENT CATERPILLAR'S
ROLEINTHEFOALDEATHSISSTILL
BEING INVESTIGATED.

Working
to Solve

the Foal
5
Mystery

  

n Kentucky Derby Saturday‘ May 5, 2001, 75 dead foals and fetuses were delivered to the

Livestock Disease Diagnostic Center at UK‘s Coldstream Research Campus on Newtown

l’ike. And while weekend deliveries of dead animals to the back door of the center is
common‘ the tremendous number that day—about 10 times more than usual~was alarming.

“I knew we had big trouble on our hands. In addition to these foals and fetuses we were aware that
too many foals in the Fayette County area were being born weak," says Lenn Harrison. veterinarian
and center director. Harrison worked alongside center pathologists practically non—stop that
weekend—the first ofmany—to try to solve the puzzle ofwhat was causing the death ofso many foals
and fetuses.

In the lab, pathologists performed a necropsy on each foal, looking at internal organs for signs of
infection as well as taking blood and tissue samples for analysis. The tissue samples offered up the clue
ofrare bacterial infections in many aborted fetuses, but the pathologists concluded that these bacteria
could not be solely responsible for the deaths. uAs the necropsies continued, our beliefgot stronger
that the bacteria were secondary to the specific cause of what was dubbed Mare Reproductive Loss
Syndrome." Harrison says.

Within the next two days the total number ofdead foals and fetuses brought to the center grew
to 276, and a team ofover 100 experts from a variety ofdisciplines in the College ongriculture joined
the investigation. A group ofhighly experienced veterinarians made up a survey to be distributed right
away to horse farms by the Kentucky 'l‘horoughbred Farm Managers Club to try to determine the
extent ofthe losses. Staffin the UK agricultural communications office also swung into action to make
certain that the industry and the public were kept abreast of the situation. And a Web site.
wywv.ukyedu/Agriculture/VetScience/mrls, was established to provide a continuing source of informa—
tion for the public.

 

6 ODYSSEY

 

 

 Eli‘s magzuine on research

ODYSSEY

__ I am happy to make a tax-deductible contribution to Odyssey.
__ Please add me to the Odyssey mailing list.

_ My address has Changed. Please send Odyssey to:

Name

 

Title

 

Company

 

Address

 

City/ State/ Zip

 

Thank you for your continuing support
of research at the University of Kentucky.

 

   

 

BUSINESS REPLY MAIL

FIRST CLASS MAIL PERMIT No.18 LEXINGTON, KY

 

 

 

POSTAGE WILL BE PAID BY ADDRESSEE

UNIVERSITY OF KENTUCKY
ODYSSEY

222 BOWMAN HALL-0059
500 S LIMESTONE
LEXINGTON KY 40508-9965

 

ND POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES

 

 

 

  

 
    
  
  

    

 

 

  

By May 10, news of the syndrome had
spread around the world: US. networks as
well as European and Middle East news
agencies were constantly calling UK for
information about this syndrome that had
the potential to disrupt the global horse
industry. Four days later, May 14, farm
managers who had returned questionnaires
confirmed the worst: the syndrome was
widespread, and it was continuing.

“At this point, a couple offaculty mem—
bers at Cluck [UK‘s Gluck Equine Re—
search Center] recalled that in 1980 and
1981 early fetal loss had occurred sponta—
neously and no cause had ever been deter-
mined,“ Harrison says. “And in both years
the losses abruptly stopped." Reported losses
during those years, however, were nowhere
near the 2001 numbers.

“A quick check ofmeteorological condi—
tions then and now,” he explains, “showed
a disturbing similarity of weather patterns
that could affect pastures where the mares
grazed." March temperatures all three years
were below normal, followed by above-
normal temperatures in April. Such a pat—
tern would result in explosive biological
activity in both plants and insects. The data
also showed frost or freeze in the third week
oprril, followed by warm temperatures a
few days later. What impact could those
factors have on the pastures?

uAlthough the convergence of weather
factors in the three years with larger than
usual abortion rates wasn’t a smoking gun,
it was suggestive that something weather—
related was responsible,“ Harrison says.

Perhaps, he thought, weather-dependent
vegetation could be linked to a variety of
possible disease-causing agents, including
mycotoxins and ergot—type alkaloids (which
are derived from fungi and can be poison—
ous to animals in high concentrations) and
phytoestrogens (compounds produced by
plants that mimic the hormone estrogen).

Tests for poisoning from mycotoxins in
feed, ergot—type alkaloids in pasture, and
phytoestrogens appeared to be negative,
but most of the tests were done after the
syndrome had occurred.

Because none of the obvious causes ap—
peared plausible, the scientists turned their
attention to still another possibility—that
the villain might be the Eastern tent cater-
pillar, whose populations were high in each
of the years in question. Harrison showed
UK agronomist Jimmy Henning some in—
formation he’d run across on these prolific
crawlers, in particular the fact that they feed
on wild cherry leaves.

And when Henning visited farms where
mares had given birth to dead or dying foals
or had aborted early—term fetuses, he dis—
covered a striking pattern: most of the
pastures had a large number ofwild cherry
trees. “I knew that cherry tree leaves carried
the precursor to a poison—naturally occur—
ring organic cyanide—that could cause
death in horses and other animals, espe—
cially cattle," says Henning. “1 also realized
that Eastern tent caterpillars, which were
known to feed on the leaves ofcherry trees
and were seemingly immune to the poison,
had been quite active during late April.”

The caterpillars became even stronger
suspects in a meeting ofveterinarians, farm
managers and the media on May 24. “At
that meeting,n Harrison says, “several vet—
erinarians confirmed that the incidence of
early fetal losses as well as stillborn and
dying foals had dramatically decreased dur—
ing the previous week. paralleling the natu—
ral decline in caterpillar numbers.”

The investigation into the possible role
played by the caterpillars continued. In
December 2001, entomology professor
Bruce Webb made an interesting discovery:
his research showed that cyanide does not
accumulate in the caterpillars, so they are
not likely to deliver appreciable amounts to

horses. This finding does not mean, how—
ever, that the caterpillars were necessarily
innocent bystanders. Research planned by
Webb this spring will investigate potential
indirect roles that the caterpillars may play.
“They produce large amounts of waste as
they feed,“ Webb explains. “Much of this
nutrient—rich material rains to the ground
under infested trees and may serve as food
for specific molds that produce toxins that
might contribute to the syndrome.”

As researchers close in on the causes of
the foal—loss syndrome, the importance of
their work is underscored by last springs
sad numbers. Between April 28 and May
12, 418 aborted equine fetuses and still-
born foals were brought to the Livestock
Disease Diagnostic Center for evaluation.
During that same period in 2000, only 60
fetuses and stillborn foals were received. On
May 23, 2001, a University of Kentucky
survey of 1 59 thoroughbred farm managers
indicated that 678, or 21 percent, of3,294
pregnant mares had experienced early fetal
loss.

So, what can be done now and in the
future to make certain that such an aberra—
tion doesn’t occur again?

Henning says that researchers have be—
gun a complex environmental monitoring
system they hope will give them advanced
warning if conditions are ripe for this to
happen again. About 13 farms are being
tested on a rotating basis, with soil, grass,
water and other samples collected, ana—
lyzed, and stored for later comparison should
foals again begin to die in such numbers.
Weather patterns are also being closely
tracked, and a massive research initiative on
the causal mechanisms of the syndrome is
planned by the College ongriculture in the
coming months.

Harrison says he hopes that the only
thoroughbred problem he will have to deal
with around Kentucky Derby time this year
is which horse to bet on.

 

For the latest information on the possible causes of Mare Reproductive Loss Syndrome,

go to www.uky.edulAgriculture/VetSciencelmrIs

 

  

 

UNIVERSITY OF KENTUCKY 7

 

 
       

  

 

 

Supercomputers

“R” Linking PCs for

Unprecedented Power

w R I T T E N B Y
ALICIA P. GREGORY

n 1993. Hank Dietz and his
graduate students at Purdue
University decided to do some
tinkering. They built a “cluster"~
took four off—the—shelf personal

 

computers, wired them together, and added their homemade network hard— RESEARCH ASSISTANT Tm MATTOX (LEFT).
ware to make a do—it-yourself supercomputer. By February 1994, the first GRAD STUDENT GALEN RASCHE AND HANK
generation oftheirspecial Linux PC clusterwas born.Itwas cheap.ltwas fast. D'ETZ W'T“ THE T°°L5 0" THE'R TRADE: A

REAR-PROJECTION SCREEN. THE CLUSTER
SUPERCOMPUTER KLATZ, AND A VIDEO WALL

And they were first.
About halfa year later, some other folks built their first cluster. named it
‘Beowulfi‘ and went down in history as the creators ofthe Linux cluster.
The Purdue group had already built several clusters by that point, the largest -
of which was named “Spareparticus.” So why didn‘t Spareparticus become
synonymous with cheap, fast. computing clusters?
“Maybe it was the name," Dietz says with a shrug. ”It just doesn‘t sound
serious enough."
In any case, he never had any intention oftreating Spareparticus as a serious
computer. \X’hat began as an experiment to test what Dietz calls a “sneaky
way” of performing certain kinds of communication within a parallel
supercomputer (work his research group had been doing since 1987) turned
into something much more.

MADE UP OF STANDARD COMPUTER MONITORS.

 

8 ODYSSEY

 

   
  

In 1993 his group had a breakthrough on
how to implement his ideas in this custom
hardware, [see sidebar at right], but they
needed to build a mock—up to try it out. “So
after a little bit ofhead—scratching, we real—
ized that instead of building the machine
from scratch the easiest way to build a
prototype would be to take a bunch ofPCs
and tie them together with a very simple
version ofour custom stuffand use Ethernet
to pass the messages between PCs," he says.

“Only after we actually built our first
cluster we realized, ‘Gee, this is workin
really well.” And by “really welln Dietz
means he was getting better performance
from the shabby—looking cluster than he was
getting from the million—dollar—plus
supercomputer that sat next to it.

And since then Dietz, who came to the
University of Kentucky in 1999, has had
two goals: one, make it easier for anyone to
build his own supercomputing cluster, and,
two, create clusters capable of high—end
processing for cutting—edge research.

Exploiting the Parallel

The concept of parallel processing can be
boiled down to a need for speed. The best
way to rev up a computer program is to
divide it into multiple fragments that can
run simultaneously, each on its own pro—
cessor (that‘s the Pentium or Athlon chip.
the brain of a PC). Intel’s Pentium 4 or
AMD‘s Athlon XP processors are based on
the same idea. Inside those tiny chips of
silicon, hundreds of things are happening
simultaneously

1

The same kind of parallelism is being

con/inner! U” page [0

 

 

Putting the “Custom”
in Network Hardware

It’s smaller than a breadbox. In fact, it’s about the size of an old-style toaster. But this
humble-looking hardware is unbelievably efficient at allowing a cluster of unmodified
PCs to do the kind of processing once possible only on pricey supercomputers.

Named PAPERS (Purdue’s Adapter for Parallel Execution and Rapid Synchroniza-
tion) and created in 1994, Dietz’s design uses a custom board that connects to PCs via
their parallel ports—the same port on your desktop computer where you normally hook
up a printer.

Why printer ports? Because the waiting period is about as short as you can get.

“If you just want to be able to get a little bit of data out of the machine as fast as possible,
the parallel printer port is the way to go, because you can get something out in about one
microsecond,” Dietz says. That’s faster than any other standard PC hardware—things
like Ethernet interfaces take a while to get started sending data.

“Standard networking hardware is very good at sending a message from one processor
to another, point—to-point. But if you want to ask questions like, ‘Is everybody done yet?’
or ‘Who’s done with their work so I can give them some more?’——the kind of questions
that come up a lot in parallel programs—you can’t answer very efficiently using point-
to—point messages.

“Those are what we call aggregate questions—they require you to accumulate data from
all of the processors.” PAPERS—which has already gone through 19 generations and
which Dietz describes as an “aggregate function network”—was specifically designed to
do this.

“In a single operation, an aggregate function network collects data from everybody,
reduces that data down to one piece of information (the answer to the question), and then
returns that to everybody.” He says if you tried to do this kind of thing with a traditional
network, there’d be a whole flurry of communications that would slow the entire system
down.

“Our network hardware is not commercially available from anybody at this time, which
means we have to build the stuff ourselves," Dietz says. “But the good news is our designs
are free, publicly available, and well documented on the Web.”

Dietz’s first acquaintance with the University of Kentucky was through Jim Lumpp,
an associate professor in electrical and computer engineering. Lumpp was working on
cloning Dietz's custom hardware for his own research about a year before Dietz came to
UK.

Dietz says the real draw of UK was the fact that it sports both a top—notch engineering
and medical school. “I got married two and a half years ago. My wife [Sabire Ozcan, a
biochemistry researcher] was in a non-tenured faculty position at Washington University
in St. Louis, I was tenured at Purdue in Indiana, and we were looking for a university
where we could both work,” he says. “There are very few universities th