Difference between revisions of "CyberShake Study 2.3"
| Line 5: | Line 5: | ||
We are planning to use Blue Waters, Stampede, and Kraken for this calculation. We plan to calculate 281 sets of 2-component SGTs for each of RWG CVM-S, RWG CVM-H, AWP CVM-S, AWP CVM-H for a total of 1124 sets of SGTs. | We are planning to use Blue Waters, Stampede, and Kraken for this calculation. We plan to calculate 281 sets of 2-component SGTs for each of RWG CVM-S, RWG CVM-H, AWP CVM-S, AWP CVM-H for a total of 1124 sets of SGTs. | ||
| − | We estimate the following requirements for each system | + | We estimate the following requirements for each system. Data estimates are for generated data we may want to keep (SGTs, seismograms, PSA). |
=== Blue Waters === | === Blue Waters === | ||
| − | Use for | + | Use for SGT calculations |
| − | 562 sets of AWP SGTs x 5000 SUs/set = ''' | + | 562 sets of AWP SGTs x 5000 SUs/set = 2.8M SUs |
| − | + | 562 sets of RWG SGTs x 5500 SUs/set = 3.1M SUs | |
| + | Total: '''5.9M SUs''' | ||
| + | 1124 sets of SGTs x 40 GB/set = '''44.0 TB''' | ||
=== Stampede === | === Stampede === | ||
| − | Use for | + | Use for RWG post-processing |
| − | |||
| − | |||
| − | |||
No numbers yet on Stampede post-processing. SGT calculations were about 4x faster on Stampede than Kraken. If we assume we can get 2x improvement: | No numbers yet on Stampede post-processing. SGT calculations were about 4x faster on Stampede than Kraken. If we assume we can get 2x improvement: | ||
| − | 562 sites x 2750 SUs/site = 1.5M SUs | + | 562 sites x 2750 SUs/site = '''1.5M SUs''' |
| − | 562 sites x 11.6 GB/site = 6.4 TB output data (seismograms, spectral acceleration) | + | 562 sites x 11.6 GB/site = '''6.4 TB''' output data (seismograms, spectral acceleration) |
Total: '''3.7M SUs''', '''28.4 TB''' | Total: '''3.7M SUs''', '''28.4 TB''' | ||
Revision as of 22:35, 20 February 2013
CyberShake Study 2.3 is a proposed study to calculate hazard curves under CyberShake 1.5 using CVM-S and CVM-H with the RWG V3.0.3 SGT code and AWP-ODC-SGT, and the Graves and Pitarka (2010) rupture variations. The goal is to calculate the same Southern California site list (281 sites) as for past CyberShake studies so we can produce comparison curves and maps, and understand the impact of the SGT codes and velocity models on the CyberShake seismic hazard.
Contents
Computational and Data Estimates
We are planning to use Blue Waters, Stampede, and Kraken for this calculation. We plan to calculate 281 sets of 2-component SGTs for each of RWG CVM-S, RWG CVM-H, AWP CVM-S, AWP CVM-H for a total of 1124 sets of SGTs.
We estimate the following requirements for each system. Data estimates are for generated data we may want to keep (SGTs, seismograms, PSA).
Blue Waters
Use for SGT calculations
562 sets of AWP SGTs x 5000 SUs/set = 2.8M SUs 562 sets of RWG SGTs x 5500 SUs/set = 3.1M SUs Total: 5.9M SUs 1124 sets of SGTs x 40 GB/set = 44.0 TB
Stampede
Use for RWG post-processing
No numbers yet on Stampede post-processing. SGT calculations were about 4x faster on Stampede than Kraken. If we assume we can get 2x improvement: 562 sites x 2750 SUs/site = 1.5M SUs 562 sites x 11.6 GB/site = 6.4 TB output data (seismograms, spectral acceleration) Total: 3.7M SUs, 28.4 TB
Kraken
Use for AWP post-processing
562 sites x 11.6 GB/site = 6.4 TB output data (seismograms, spectral acceleration) 562 sites x 5500 SUs/site = 3.1M SUs
SCEC storage
1124 sites x 11.6 GB/site = 12.7 TB stored output data 1124 sites x 4.9 GB/site = 5.4 TB workflow logs
