Presented by Scientific Data Management Center Nagiza F. Samatova Network and Cluster Computing Computer Sciences and Mathematics Division
Jan 05, 2016
Presented by
Scientific Data Management Center
Nagiza F. SamatovaNetwork and Cluster Computing
Computer Sciences and Mathematics Division
2 Samatova_SDM_0611
Scientific Data Management Center
SciDAC Review, Issue 2, Fall 2006 Illustration: A. Tovey
Lead Institution: LBNL
PI: Arie Shoshani
Laboratories:ANL, ORNL, LBNL, LLNL, PNNL
Universities:NCSU, NWU, SDSC, UCD, U. Utah
Established 5 years ago (SciDAC-1)
Successfully re-competed for next 5 years (SciDAC-2)
Featured in Fall 2006 issue of SciDAC Review magazine
3 Samatova_SDM_0611
SDM InfrastructureUses three-layer organization of technologies
Scientific Process Automation (SPA)
Data Mining and Analysis (DMA)
Storage Efficient Access (SEA)
Operating system
Hardware (e.g., Cray XT3, IBM Blue/Gene L)
Operating system
Hardware (e.g., Cray XT3, IBM Blue/Gene L)
Integrated approach:• To provide a scientific
workflow capability
• To support data mining and analysis tools
• To accelerate storage and access to data
Benefits scientists by• Hiding underlying parallel
and indexing technology
• Permitting assembly of modules using workflow description tool
Goal: Reduce data management overhead
4 Samatova_SDM_0611
Automating scientific workflow in SPA enables mining of diverse biology databases
Scientific discovery is a multi-step process. SPA-Kepler workflow system automates and manages this process.
Illustration: A. Tovey
Tasks that required hours or days can now be completed in minutes, allowing biologists to spend their time saved on science
Contact: Terence Critchlow, LLNL ([email protected])
5 Samatova_SDM_0611
Data analysis for fusion plasma
Plot of orbits in cross-section of a fusion experiment shows different types of orbits, including circle-like “quasi-periodic orbits” and “island orbits.” Characterizing the topology of orbits is challenging, as experimental and simulation data are in the form of points rather than a continuous curve. We are successfully applying data mining techniques to this problem.
Feature selection techniques used to identify key parameters relevant to the presence of edge harmonic oscillations in the DIII-D tokomak.
Contact: Chandrika Kamath, LLNL ([email protected])
Err
or
0.16
0.18
0.2
0.22
0.24
0.26
0.28
0.3
0 5 10 15 20 25 30 35
Features
PCA FilterDistance
ChiSquareStump
Boosting
6 Samatova_SDM_0611
Searching and indexing with FastBit Gleaning insights about combustion simulation
About FastBit:
• Extremely fast search of large databases
• Outperforms commercial software
• Used by various applications: combustion, STAR, astrophysics visualization
Collaborators:
SNL: J. Chen, W. Doyle
NCSU: T. Echekki
Illustration: A. ToveyIllustration: A. Tovey
Finding & tracking of combustion flame frontsFinding & tracking of combustion flame fronts
Contact: John Wu, LBNL ([email protected])
Searching for regions that satisfy particular criteria is a challenge. FastBit efficiently finds regions of interest.
7 Samatova_SDM_0611
Searches in STAR experiment Finding one event out of a million
Illustration: A. Tovey
STAR scientists extract data of interest within 15 minutes. Accomplishing the same task before consumed weeks.
FastBit and Storage Resource Manager enable efficient mining of hundreds of terabytes of data generated by STAR detector
Contact: John Wu, LBNL ([email protected])
8 Samatova_SDM_0611
Parallel statistical computing with pRGoal: Provide scalable high-performance statistical data analysis framework to help scientists perform interactive analyses of produced data to extract knowledge
• Able to use existing high-level (i.e., R) code
• Requires minimal effort for parallelizing
• Offers identical interface
• Provides efficient and scalable performance
Contact: Nagiza Samatova, ORNL ([email protected])
9 Samatova_SDM_0611
Parallel input/outputScaling computational science
Multi-layer parallel I/O design:
Supports Parallel-netCDF library built on top of MPI-IO implementation calledROMIO, built in turn on top of Abstract Device Interface for I/O system, used to access parallel storage system.
Benefits to scientists:• Brings performance, productivity,
and portability
• Improves performance by order of magnitude
Orchestration of data transfers and speedy analyses depends on efficient systems for storage, access, and movement of data among modules.
Illustration: A. ToveyIllustration: A. Tovey
Contact: Rob Ross, ANL ([email protected])
10 Samatova_SDM_0611
Speeding data transfer with PnetCDF
P0P0 P1P1 P2P2 P3P3
netCDFnetCDF
Parallel File SystemParallel File System
Parallel netCDFParallel netCDF
P0P0 P1P1 P2P2 P3P3
Parallel File SystemParallel File System
Illustration: A. ToveyIllustration: A. Tovey
Rate of data transfer using HDF5 decreases when a particular problem is divided among more processors. In contrast, parallel version of netCDF improves because of low-overhead nature of PnetCDF and its tight coupling to MPI-IO.
Enables high performance parallel I/O to netCDFdata sets
Achieves up to 10-fold performance improvement over HDF5
Contact: Rob Ross, ANL ([email protected])
Inter-process communication
11 Samatova_SDM_0611
Contact
Nagiza SamatovaNetwork and Cluster ComputingComputer Science and Mathematics Division(865) [email protected]
11 Samatova_SDM_0611