| |
|
|
|
|
 |
NCSA NEWS |
|
|
|
|
High-throughput applications, in contrast to
high-performance applications, rely not on the
computer's ability to complete a computation
faster than ever, but on its ability to execute
massive volumes of computations in a fixed
amount of time, measured in days or weeks rather
than nanoseconds. Using computers to generate
behavioral data on complex phenomena (e.g.
studying a biological system, testing the
design of a new hardware component, or evaluating the
risk of an investment) researchers with
high-throughput problems can never have too much data.
For such projects the question is: How much
data can I generate by my deadline so that the
reliability of my model is maximized?
It is a question of throughput. Symera's answer
is to keep an entire domain of Windows NT
workstations under surveillance, apprehending
the idle cycles of personal computers and
applying them, collectively, to solving
high-throughput problems. It may be able to provide an
alternative for researchers wishing to maximize the
resources applied to the solution of their
problem.
|
| |
