Last update:
Apr 1, 2000
|
|
|
Design and First Tests of the CDF Run 2 Farms
|
Jaroslav Antos1,
Mark Breitung2,
Tony Chan3,
Paoti Chang3,
Yen-Chu Chen3,
Troy Dawson2,
Jim Fromm2,
Lisa Giacchetti2,
Tanya Levshina2,
Igor Mandrichenko2,
Ray Pasetes2,
Marilyn Schweitzer2,
Karen Shepelak2,
Miroslav Siket4,
Dane Skow2,
Stephen Wolbers2,
G.P. Yeh2,
Ping Yeh3
- Slovak Academy of Science
- Fermilab
- IHEP, Academia Sinica, Taiwan
- Comenium University, Slovak Republic
Speaker:
Stephen Wolbers
CDF will write data in Run 2 at twenty times the rate of Run 1. The
offline production must be able to handle this rate. Components of that
production include a large PC farm, I/O systems to read/write data
to/from mass storage, and a system to split the reconstructed data into
physics streams which are required for analysis. All of the components
must work together as seamlessly as possible to ensure the necessary
throughput.
A description will be given of the overall hardware and software design
for the system. A small prototype farm has been used for about one year
to study performance, to test software designs and is used for the first
Mock Data Challenge. Results from the tests will be shown and experience
from the first Mock Data Challenge will be discussed. The hardware for
the first production farm is in place and will be used for the second
Mock Data Challenge. Finally, the possible scaling of the system to
handle even larger rates foreseen later in Run 2 will be described.
Presentation: |
|