2012 02 pre_hbs_grid_overview_ianstokesrees_pt1
1 like392 views
The document discusses Ian Stokes-Rees' presentation on grid computing. It provides an overview of how grid computing originated in high energy physics and has expanded to life sciences research. Stokes-Rees discusses current US cyberinfrastructure including XSEDE and the Open Science Grid, as well as how grids support science portals, data access, user credentials, and security. The presentation covers how grids help address the large data and computing needs of projects in fields like high energy physics experiments, protein structure determination, cryo electron microscopy, and molecular dynamics simulations.
2012 02 pre_hbs_grid_overview_ianstokesrees_pt1
- 1. Grid Computing Overview Federating Compute and Storage Resources to Accelerate Research and Aid Collaboration Ian Stokes-‐Rees, PhD Harvard, Cambridge, USA http://ac.seas.harvard.edu ijstokes@seas.harvard.edu
- 2. Slides and Contact ijstokes@seas.harvard.edu http://linkedin.com/in/ijstokes http://slidesha.re/ijstokes-hbs2012_pt1 http://slidesha.re/ijstokes-hbs2012_pt2 http://ac.seas.harvard.edu http://www.opensciencegrid.org http://www.xsede.org Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 3. Today Grid Computing Origins High Energy Physics Life Science Computing Survey US Cyberinfrastructure XSEDE Open Science Grid Science Portals Data Access User Credentials Security Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 4. Things ISummary About ’m Excited Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 5. About Me Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 6. High Energy Physics Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 7. Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 8. Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 9. Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 10. 40 MHz bunch crossing rate 10 million data channels 1 KHz level 1 event recording rate 1-10 MB per event 14 hours per day, 7+ months / year 4 detectors 6 PB of data / year globally distribute data for analysis (x2) Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 11. Data and Compute Intensive Life Sciences
- 12. Study of Protein Structure and Function 400m 1mm 10nm • Shared scientiLic data collection facility • Data intensive (10-‐100 GB/day) Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 13. Cryo Electron Microscopy • Previously, 1-10,000 images, managed by hand • Now, robotic systems collect millions of hi-res images • estimate 250,000 CPU-hours to reconstruct model Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu
- 14. Molecular Dynamics Simulations 1 fs time step 1ns snapshot 1 us simulation 1e6 steps 1000 frames 10 MB / frame 10 GB / sim 20 CPU-years 3 months (wall- clock) Grid Overview - Ian Stokes-Rees ijstokes@seas.harvard.edu