Tera Grid Gateway User Concept Supporting Users

Tera. Grid Gateway User Concept – Supporting Users V. E. Lynch, M. L. Chen, J. W. Cobb, J. A. Kohl, S. D. Miller, S. S. Vazhkudai Oak Ridge National Laboratory In collaboration with many teams: NSTG, SNS Scientific Computing, Mc. Stas group, Open Science Grid, Tech-X Corp, and the Tera. Grid Partners teams.

What is a Science Gateway? · A Science Gateway - Enables scientific communities of users with a common scientific goal - Uses high performance computing - Has a common interface - Leverages community investment · Three common forms: - Web-based Portals - Application programs running on users' machines but accessing services in Tera. Grid - Coordinated access points enabling users to move seamlessly between Tera. Grid and other grids.

How can a Gateway help? · Make science more productive - Researchers use same tools Complex workflows Common data formats Data sharing · Bring Tera. Grid capabilities to the broad science community - Lots of disk space Lots of compute resources Powerful analysis capabilities A nice interface to information

What is the Tera. Grid? 20 computers at 11 facilities 10 Gbps network Over a petaflop of computing power UW 136, 470 CPU-cores 60 petabytes long-term storage UC/ANL NCAR Growing NCSA Caltech GIG PSC PU UNC/RENCI IU ORNL USC/ISI SDSC LONI TACC Resource Provider (RP) Software Integration Partner Grid Infrastructure Group (UChicago) (GIG) 4 NICS

Neutron Science Tera. Grid Gateway · Focus is neutron science · Connects facilities with cyberinfrastructure · Bridges cyberinfrastructure · Combines Tera. Grid computational resources with neutron datasets · Data movement across Tera. Grid · Outreach to neutron science 5

Community Certificate and Account · Gateways with community accounts scale to thousands of facility users · Have Jimmy Neutron community accounts on 14 Tera. Grid computers · Use Jimmy Neutron Community Certificate from SNS community account · Record end-user identification for auditing and return of results 6 Community Account

Before Gateways · Large facilities: - Recorded histogram data from experiments · Users: - Took their data home on floppy disk in pocket Saved permanent copy on hard disk Did not have event data to change histogram Translated data into format needed for analysis Wrote their own code to read analyze data Installed discipline focused software on their PC for analysis Installed plotting programs/libraries to plot analysis output 7 Tera. Grid

After Gateways · Large facilities: - Record data from multiple facilities · - · SNS, HFIR, LENS, IPNS, LUJAN, … Save permanent copy of raw data Bin event data into histogram Translate data into standard Ne. Xus format Have analysis and simulation programs available from portal Use remote Tera. Grid cyberinfrastructure for computations Have visualization capability in portal Users: - Use portal from web for all data, analysis, and visualization 8 Tera. Grid

Gateway Savings · Users do not duplicate efforts · Facilities do not duplicate efforts · Data is not lost · Data is easily shared · Natural way to integrate community contributed instrument specific software for hosting and wide availability to many facility users · Analysis is done quickly on high performance computers 9 Tera. Grid

Portal submit to Tera. Grid 10 Tera. Grid

Simulation Service · Simulation of neutron instrument is available in portal with Mc. Stas · Simulations agree with experimental results · Linear scaling to 1024 cores · Output is Ne. Xus · Use cases: - Instrument design and construction - Experiment planning and analysis 11 Tera. Grid

Fitting Service • Fits theoretical models to the Ne. Xus data files from the experiments • Adaptive nonlinear least squares algorithm implemented in parallel • Linear speedup to 32 cores • Service to run on Tera. Grid 12 Tera. Grid

Reduction Service • Reduction software is available tor backscattering and reflectometry through portal • Calculations will be sent to local cluster and Tera. Grid • Attempted to parallelize this calculation by distributing regions of Backscattering Data the time-of-flight to each processor. Reduction • Each processor read only its region of the Ne. Xus input data file and write a new file containing only that region. • Each processor performs the data reduction on its file. • The results are merged at the end of the calculation 13 Tera. Grid Reflectometry Data Reduction

Job Information Service • Portal Job • • 14 Tera. Grid Information Service tells where jobs is running, when it started and status Daily tests of submitting five simultaneous remote portal jobs Percentage success is > 82%

Tests of Remote Job Submission • Difficult to diagnose the • 15 Tera. Grid problem from the Globus output. - Check the status of the computer - Look at the output files Some problems diagnosed: - Updated software on a computer that required relinked executables - Globus software setting the wrong time limit - Batch prologue script that killed jobs on same core - Long queue waits - Firewall installed

Conclusions · Gateways help facilities scale to a large number of users · Gateways give facilities access to high performance computing such as the Tera. Grid · Gateways enable a scientific community to use community software through a common interface · Researchers are more productive if they use the same tools, use a common data format, and share data easily 16 Tera. Grid