german astrophysical virtual observatory
DESCRIPTION
German Astrophysical Virtual Observatory. Gadget on GAVOGRID. GAVO Architecture. VO-Sites. Grid Enabled Sites. GAVO WebPortal. HTTP, SOAP. Secure Protocols. WebServices. GridServices. Temporary DataSpace. UserSpace. ConeSearch. Matcher. Files. DB. Computing Tasks. - PowerPoint PPT PresentationTRANSCRIPT
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
German Astrophysical Virtual Observatory
Gadget on GAVOGRID
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
WebServices GridServices
GAVO Architecture
GAVO WebPortal
ConeSearch
Files DB
HTTP, SOAP
MatcherTemporaryDataSpace
UserSpace
Computing Tasks
Secure Protocols
Grid Enabled Sites
Normal security realm
Grid security realm
VO-Sites
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GAVO-Grid
GAVO-GRID (testbed) using Globus
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GAVO-GRID
Testcase for GAVO:running MPI-application across site-boundaries:
Gadget is a widely used particle-code for cosmological simulations
Gadget (MPI-version) requires 2n processors
Gadget roughly uses for each timestep80% calculation, 20% communication
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GAVO-GRID
Running MPI-applications with Globus / mpichG2
- all nodes communicate with each other(ip-adresses known)
- using ssh instead of rsh
- requires a range of TCP/IP-ports open(globus-tcp-range)
- allocation of bandwidth
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GAVO-GRID
Results from tests with
-uniform nodes
Wall time
Comm time
On board 167,05 0,11Across site 179,69 1,05(np = 8) 105,3 1,30
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GAVO-GRID
Results from tests with heterogenous nodes
Np Wall time Comm time
On site 2/2 AMD/ XEON
162,03 0,11
Across site 4 AMD
97,7 1,16
On site 4/4 AMD/XEON
85,9 0,15
Across site 4/4 AMD/XEON
105,33 1,18
Across site 8/8 AMD/XEON
77,63 1,96
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GRID-Collaboration
Simulations for cosmological problems improve with the number of particles.
There is a clear relation between available computing ressources and the number of particles a simulation can handle.
Integration of clusters using grid-technology is an optionwhich can significantly lower the costs of getting biggersimulations running.
One of the main issues then is allocation of more bandwidth
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GRID-Collaboration in Potsdam:
Potsdam has a good net infrastructure, where more bandwidth is relatively easy available
For testing purposes, a low cost solution providing more bandwith could be installed, including
AEI, AIP, PIK, Uni Potsdam
This could be a first step of building part of the D-Grid at Potsdam.
AEI-Clustertag, 12.05.2004 GAVO and GridComputing
GRID-Collaboration
AEI and AIP are using grid-technologydo demanding cosmological simulations
working towards a grid-integration of their respective clusters
working on specific problems (visualization of simulation results, distributed file systems)