SlideShare a Scribd company logo
1 of 29
Download to read offline
National Technical University of Athens
School of Civil Engineering
A shared-filesystem-memory approach for
running IDA in parallel over informal
computer clusters
D. Vamvatsikos
National Technical University of Athens
…and friends….
EOSD 2017 Opensees Days Europe Porto, June 19-20, 2017
2
Introduction
• Seismic performance evaluation. How?
• Incremental Dynamic Analysis is accurate but slow
– Multiple nonlinear dynamic analyses, Multiple records
– 1 CPU  Needs patience!
• Can we run it faster?
– Use N CPUs.
– Parallelize each nonlinear analysis? OpenSeesMP
– Parallelize IDA? Easier, repetitive, more efficient 
3
LA9 building
• 9 stories, 1 basement, T1 = 2.3s
• 2D model with internal gravity frame, P-Delta, beam-hinges
4
The IDA tracing problem
• High variability  many
records
• Complex shapes  many runs
• Focus: 20 recs, 12 runs each
• One i5 core = 40 hours
• Only for academic use!
5
How to run IDA in parallel?
• Each record is completely independent from the others
– Distribute single-record IDA tasks
– Use up to 20 cores, pure linear efficiency
• Within each records the runs are not independent
– Distribute single runs  need new tracing algorithm
– Up to 20x12=240 cores (ideally)
– Dependencies  Runs may be wasted
6
Target Application Environment
• MATLAB + OpenSees
• Informal cluster of dissimilar processors
– Multiple cores per physical processor
– New and old PCs
– Unreliable networks
– Incompetent system admins
– Random PC deaths (Win10 reboots, student magic….)
• Need resilience over an unreliable “cluster”
7
Scheme 1: Distribute records
• Shared memory approach
– Everything stored in a file server (shared directory)
– One job = One .mat file
– Race conditions? Use directories as lockfiles
• Master / slave model
– Master assigns tasks (records)
– Slaves run task, return one IDA curve per record
– Master assigns tasks to self  Max efficiency
Master CPU
Slave CPUs
Send record
Receive IDA curve
Assign record to self
8
Scheme 1: Coarse grained
• Pros
– Negligible communication overhead
– Uses existing IDA tracing algorithms (hunt & fill)
– Easy programming
• Cons
– Low scalability
– Records / CPUs <> integer  CPUs are idling
e.g. 20 records / 3 cpus  2 cpus run 7 records, one runs 6.
9
Scheme 2: Distribute records then runs
• Master / slave-1 / slave-2 model
– Master assigns tasks (records)
– Slaves-1 assign single runs to slaves-2
– Master & slaves-1 assign runs to selves
Master CPU
Slave-1 CPUs
Send record
Receive IDA curve
Assign record/runs to self
Assign runs to self
Slave-2 CPUs
Send run IM
Receive EDPs
10
Scheme 2: Medium-grained
• Pros
– Still low communication overhead
– Excellent scalability
– Almost linear efficiency
– Minimal CPU idling
• Cons
– Tougher to program: Dynamic Allocation of Tasks
– Slave-cores compete for jobs (first-come, first-serve)
– Needs new IDA tracing algorithm
11
Serial Hunt & Fill IDA tracing
• Hunt-up (large steps)
• Bisect to find collapse
• Fill the gaps going down
• Hunt-up and Bisect are
unpredictable due to
collapse
Hunt-up
Fill-down
Bisect
12
Parallel Hunt & Fill IDA tracing
• 3 CPUs  more non-
converged runs
• Fill-in is still efficient
• 14 runs instead of 12
• More CPUs in hunt-up or
Bisect  more waste!
• Still, more efficient that
stepping algorithm
Hunt-up
Fill-down
Bisect
13
Example: 20 records, 3 identical CPUs
• Parallel hunt&fill achieves near linear performance (2.96/3 = 99%)
• When the last 2 records were run, cpu3 was helping alternatively cpus 1,2
14
Performance Comparison
• For 60 CPUs, each record uses 3+ cores simultaneously
• Efficiency drops to 88% only
How is this done in software?
15
• Master script:
[anls]=runIDA_NDpx(anls,trace,fupdate,sharedmem,runmode);
• Slave script
[icase]=runIDA_NDpx_slave(sharedmem,masterflag,...,imultcpu);
• One record subscript
[icase]=runIDA_NDpx_slave_onerec(sharedmem,...,imultcpu);
• One run subscript
[icase]=runIDA_NDpx_slave_onerun(sharedmem,...,imultcpu);
• Subscripts are run based on availability of jobs
Analysis and tracing
parameters
Tcl file updating info
for all runs & models
Shared memory
IP address
0 = partition records
1 = partition runs
Assign a number
to each core
Set to 1 if master
self-assigned the run
16
Application: Model parameter uncertainties
• Monte-Carlo based
– Use full IDA (Ibarra, Dolsek, Vamva & Frag)
– Approximate IDA
• Response Surface (Liel et al)
• SPO2IDA (Frag & Vamva) + …
• Moment-Estimation based
– FOSM (Ibarra, Liel et al, Lee & Mosalam, Vamva & Frag)
– PEM (Vamva & Frag) + …
17
Beam point-hinge model
• Allows hardening & softening with pinching loops
• Residual plateau terminates at ultimate rotation  True hinge
• Means: ah = 10%, μc = 3, ac = -50%, r = 50%, μu = 6, aMy = 1
18
“Mean” model IDA curves
• 30 records, 12 runs each
• P-Delta and beam hinging 
Clear flatlines
• Summarization into fractiles
• Med. collapse capacity
• Large dispersion
19
200 median IDAs (real or approximate)
20
Let’s try to improve
• Progressive application of LHS
– Initial small sample
– Double size by adding samples in each step
– Stop when adequate accuracy
• Allow sampling on a record-by-record basis
– Old  full multi-record IDA for each parameter sample
– New  single-record IDA for each sample
– Recycle records if not enough
– Not new ideas, e.g. Schotanus & Franchin
21
Progressive LHS (1)
• Start with Npoints ≥ Nvars
• Say 4 samples for 2 vars
• Run analysis
• Double the samples  8
22
Progressive LHS (1)
• Start with Npoints ≥ Nvars
• Say 4 samples for 2 vars
• Run analysis
• Double the samples  8
23
Progressive LHS (2)
• Run analysis
• Double the samples  16
• Double again  32
• Stop, e.g. when dispersion is
“stable”
24
Good but not perfect
• Pros
– Excellent scalability
– Removes problem of a priori definition of sample size
– Handle large number of r.v’s: insignificants disappear!
– Better coverage of sample space (like orthogonal LHS)
• Cons
– Cannot use fast Iman-Conover algorithm for correlation
– Prefer genetic-algorithms, e.g. Charmpis&Panteli (2004)
– These may be slower but more accurate
25
Include records in LHS sampling
• If not enough records: Recycle + vary incident angle
No X1 X2 … XN XN+1 XN+2
1 x1,1 x1,2 … x1,N ang1 Rec1
2 x2,1 x2,2 … x2,N ang2 Rec2
… … … … … … …
M xM,1 xM,2 … xM,N angM RecM
M+1 xM+1,1 xM+1,2 … xM+1,N angM+1 Rec1
M+2 xM+2,1 xM+2,2 … xM+2,N angM+2 Rec2
… … … … … … ….
26
Good but not perfect (again)
• Pros
– Handles large number of random vars ( > 500)
– Can vary incident angle (More records is better though)
– Place influential vars first to better capture correlation
• Cons
– Cannot use some “fast-IDA” techniques (e.g. Liel et al response
surface, or Azarbakht-Dolsek priority lists, or SPO2IDA)
– Cannot distinguish epistemic from aleatory (Do we care?)
27
Convergence
• Stable after 4th generation
• 160 samples
• Stable after 5th generation
•  320 samples
• Note 270 random vars!
28
Compare with “mean” model
• Medians differ!
• Conservative bias due to
correlation structure
• Betas are similar
• In contrast with other results
• Still needs work!
29
Some concluding remarks
• Is this worth it?
– Why not OpenSeesSP / OpenSeesMP?
– Why not HTCondor or similar?
– Why Matlab and not Python?
• Where to go now?
– If found useful, easily ported to Python
– Works for MSA, IDA, Cloud, any strategy
– You can use this now: Will release by end of summer!

More Related Content

What's hot

FlinkML: Large Scale Machine Learning with Apache Flink
FlinkML: Large Scale Machine Learning with Apache FlinkFlinkML: Large Scale Machine Learning with Apache Flink
FlinkML: Large Scale Machine Learning with Apache FlinkTheodoros Vasiloudis
 
Practical RISC-V Random Test Generation using Constraint Programming
Practical RISC-V Random Test Generation using Constraint ProgrammingPractical RISC-V Random Test Generation using Constraint Programming
Practical RISC-V Random Test Generation using Constraint Programminged271828
 
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...MLconf
 
Transformer Mods for Document Length Inputs
Transformer Mods for Document Length InputsTransformer Mods for Document Length Inputs
Transformer Mods for Document Length InputsSujit Pal
 
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.GeeksLab Odessa
 
強化学習の分散アーキテクチャ変遷
強化学習の分散アーキテクチャ変遷強化学習の分散アーキテクチャ変遷
強化学習の分散アーキテクチャ変遷Eiji Sekiya
 
Suneel Marthi - Deep Learning with Apache Flink and DL4J
Suneel Marthi - Deep Learning with Apache Flink and DL4JSuneel Marthi - Deep Learning with Apache Flink and DL4J
Suneel Marthi - Deep Learning with Apache Flink and DL4JFlink Forward
 
Flink Gelly - Karlsruhe - June 2015
Flink Gelly - Karlsruhe - June 2015Flink Gelly - Karlsruhe - June 2015
Flink Gelly - Karlsruhe - June 2015Andra Lungu
 
Requirements in Cyber-Physical Systems: Specifications and Applications
Requirements in Cyber-Physical Systems: Specifications and ApplicationsRequirements in Cyber-Physical Systems: Specifications and Applications
Requirements in Cyber-Physical Systems: Specifications and ApplicationsLionel Briand
 
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...Kalman Graffi
 
Ufuc Celebi – Stream & Batch Processing in one System
Ufuc Celebi – Stream & Batch Processing in one SystemUfuc Celebi – Stream & Batch Processing in one System
Ufuc Celebi – Stream & Batch Processing in one SystemFlink Forward
 
Flink Forward SF 2017: Dean Wampler - Streaming Deep Learning Scenarios with...
Flink Forward SF 2017: Dean Wampler -  Streaming Deep Learning Scenarios with...Flink Forward SF 2017: Dean Wampler -  Streaming Deep Learning Scenarios with...
Flink Forward SF 2017: Dean Wampler - Streaming Deep Learning Scenarios with...Flink Forward
 
Gelly in Apache Flink Bay Area Meetup
Gelly in Apache Flink Bay Area MeetupGelly in Apache Flink Bay Area Meetup
Gelly in Apache Flink Bay Area MeetupVasia Kalavri
 
Tech Days 2015: User Presentation Vermont Technical College
Tech Days 2015: User Presentation Vermont Technical CollegeTech Days 2015: User Presentation Vermont Technical College
Tech Days 2015: User Presentation Vermont Technical CollegeAdaCore
 
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big Computing
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big ComputingEuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big Computing
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big ComputingJonathan Dursi
 
Object Detection with Tensorflow
Object Detection with TensorflowObject Detection with Tensorflow
Object Detection with TensorflowElifTech
 
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...Spark Summit
 
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15MLconf
 
MODELS 2019: Querying and annotating model histories with time-aware patterns
MODELS 2019: Querying and annotating model histories with time-aware patternsMODELS 2019: Querying and annotating model histories with time-aware patterns
MODELS 2019: Querying and annotating model histories with time-aware patternsAntonio García-Domínguez
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer FarooquiDatabricks
 

What's hot (20)

FlinkML: Large Scale Machine Learning with Apache Flink
FlinkML: Large Scale Machine Learning with Apache FlinkFlinkML: Large Scale Machine Learning with Apache Flink
FlinkML: Large Scale Machine Learning with Apache Flink
 
Practical RISC-V Random Test Generation using Constraint Programming
Practical RISC-V Random Test Generation using Constraint ProgrammingPractical RISC-V Random Test Generation using Constraint Programming
Practical RISC-V Random Test Generation using Constraint Programming
 
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...
Avi Pfeffer, Principal Scientist, Charles River Analytics at MLconf SEA - 5/2...
 
Transformer Mods for Document Length Inputs
Transformer Mods for Document Length InputsTransformer Mods for Document Length Inputs
Transformer Mods for Document Length Inputs
 
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.
Java/Scala Lab 2016. Сергей Моренец: Способы повышения эффективности в Java 8.
 
強化学習の分散アーキテクチャ変遷
強化学習の分散アーキテクチャ変遷強化学習の分散アーキテクチャ変遷
強化学習の分散アーキテクチャ変遷
 
Suneel Marthi - Deep Learning with Apache Flink and DL4J
Suneel Marthi - Deep Learning with Apache Flink and DL4JSuneel Marthi - Deep Learning with Apache Flink and DL4J
Suneel Marthi - Deep Learning with Apache Flink and DL4J
 
Flink Gelly - Karlsruhe - June 2015
Flink Gelly - Karlsruhe - June 2015Flink Gelly - Karlsruhe - June 2015
Flink Gelly - Karlsruhe - June 2015
 
Requirements in Cyber-Physical Systems: Specifications and Applications
Requirements in Cyber-Physical Systems: Specifications and ApplicationsRequirements in Cyber-Physical Systems: Specifications and Applications
Requirements in Cyber-Physical Systems: Specifications and Applications
 
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...
IEEE P2P 2013 - Bootstrapping Skynet: Calibration and Autonomic Self-Control ...
 
Ufuc Celebi – Stream & Batch Processing in one System
Ufuc Celebi – Stream & Batch Processing in one SystemUfuc Celebi – Stream & Batch Processing in one System
Ufuc Celebi – Stream & Batch Processing in one System
 
Flink Forward SF 2017: Dean Wampler - Streaming Deep Learning Scenarios with...
Flink Forward SF 2017: Dean Wampler -  Streaming Deep Learning Scenarios with...Flink Forward SF 2017: Dean Wampler -  Streaming Deep Learning Scenarios with...
Flink Forward SF 2017: Dean Wampler - Streaming Deep Learning Scenarios with...
 
Gelly in Apache Flink Bay Area Meetup
Gelly in Apache Flink Bay Area MeetupGelly in Apache Flink Bay Area Meetup
Gelly in Apache Flink Bay Area Meetup
 
Tech Days 2015: User Presentation Vermont Technical College
Tech Days 2015: User Presentation Vermont Technical CollegeTech Days 2015: User Presentation Vermont Technical College
Tech Days 2015: User Presentation Vermont Technical College
 
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big Computing
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big ComputingEuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big Computing
EuroMPI 2016 Keynote: How Can MPI Fit Into Today's Big Computing
 
Object Detection with Tensorflow
Object Detection with TensorflowObject Detection with Tensorflow
Object Detection with Tensorflow
 
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...
Ernest: Efficient Performance Prediction for Advanced Analytics on Apache Spa...
 
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15
Narayanan Sundaram, Research Scientist, Intel Labs at MLconf SF - 11/13/15
 
MODELS 2019: Querying and annotating model histories with time-aware patterns
MODELS 2019: Querying and annotating model histories with time-aware patternsMODELS 2019: Querying and annotating model histories with time-aware patterns
MODELS 2019: Querying and annotating model histories with time-aware patterns
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 

Similar to A shared-filesystem-memory approach for running IDA in parallel over informal computer clusters

OpenMP tasking model: from the standard to the classroom
OpenMP tasking model: from the standard to the classroomOpenMP tasking model: from the standard to the classroom
OpenMP tasking model: from the standard to the classroomFacultad de Informática UCM
 
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...John Gunnels
 
Java Thread and Process Performance for Parallel Machine Learning on Multicor...
Java Thread and Process Performance for Parallel Machine Learning on Multicor...Java Thread and Process Performance for Parallel Machine Learning on Multicor...
Java Thread and Process Performance for Parallel Machine Learning on Multicor...Saliya Ekanayake
 
On Extending MapReduce - Survey and Experiments
On Extending MapReduce - Survey and ExperimentsOn Extending MapReduce - Survey and Experiments
On Extending MapReduce - Survey and ExperimentsYu Liu
 
11-risc-cisc-and-isa-w.pptx
11-risc-cisc-and-isa-w.pptx11-risc-cisc-and-isa-w.pptx
11-risc-cisc-and-isa-w.pptxSuma Prakash
 
Nafems15 systeme
Nafems15 systemeNafems15 systeme
Nafems15 systemeSDTools
 
04 accelerating dl inference with (open)capi and posit numbers
04 accelerating dl inference with (open)capi and posit numbers04 accelerating dl inference with (open)capi and posit numbers
04 accelerating dl inference with (open)capi and posit numbersYutaka Kawai
 
Spil Storage Platform (Erlang) @ EUG-NL
Spil Storage Platform (Erlang) @ EUG-NLSpil Storage Platform (Erlang) @ EUG-NL
Spil Storage Platform (Erlang) @ EUG-NLThijs Terlouw
 
Crash course on data streaming (with examples using Apache Flink)
Crash course on data streaming (with examples using Apache Flink)Crash course on data streaming (with examples using Apache Flink)
Crash course on data streaming (with examples using Apache Flink)Vincenzo Gulisano
 
Spark Summit EU talk by Berni Schiefer
Spark Summit EU talk by Berni SchieferSpark Summit EU talk by Berni Schiefer
Spark Summit EU talk by Berni SchieferSpark Summit
 
Nafems15 Technical meeting on system modeling
Nafems15 Technical meeting on system modelingNafems15 Technical meeting on system modeling
Nafems15 Technical meeting on system modelingSDTools
 
A Lightweight Infrastructure for Graph Analytics
A Lightweight Infrastructure for Graph AnalyticsA Lightweight Infrastructure for Graph Analytics
A Lightweight Infrastructure for Graph AnalyticsDonald Nguyen
 
Hadoop performance optimization tips
Hadoop performance optimization tipsHadoop performance optimization tips
Hadoop performance optimization tipsSubhas Kumar Ghosh
 
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...zionsaint
 
Top schools in gudgao
Top schools in gudgaoTop schools in gudgao
Top schools in gudgaoEdhole.com
 
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...EUDAT
 
What is Distributed Computing, Why we use Apache Spark
What is Distributed Computing, Why we use Apache SparkWhat is Distributed Computing, Why we use Apache Spark
What is Distributed Computing, Why we use Apache SparkAndy Petrella
 
Top schools in noida
Top schools in noidaTop schools in noida
Top schools in noidaEdhole.com
 

Similar to A shared-filesystem-memory approach for running IDA in parallel over informal computer clusters (20)

Parallel Algorithms
Parallel AlgorithmsParallel Algorithms
Parallel Algorithms
 
OpenMP tasking model: from the standard to the classroom
OpenMP tasking model: from the standard to the classroomOpenMP tasking model: from the standard to the classroom
OpenMP tasking model: from the standard to the classroom
 
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...
Making_Good_Enough...Better-Addressing_the_Multiple_Objectives_of_High-Perfor...
 
Ch13.pdf
Ch13.pdfCh13.pdf
Ch13.pdf
 
Java Thread and Process Performance for Parallel Machine Learning on Multicor...
Java Thread and Process Performance for Parallel Machine Learning on Multicor...Java Thread and Process Performance for Parallel Machine Learning on Multicor...
Java Thread and Process Performance for Parallel Machine Learning on Multicor...
 
On Extending MapReduce - Survey and Experiments
On Extending MapReduce - Survey and ExperimentsOn Extending MapReduce - Survey and Experiments
On Extending MapReduce - Survey and Experiments
 
11-risc-cisc-and-isa-w.pptx
11-risc-cisc-and-isa-w.pptx11-risc-cisc-and-isa-w.pptx
11-risc-cisc-and-isa-w.pptx
 
Nafems15 systeme
Nafems15 systemeNafems15 systeme
Nafems15 systeme
 
04 accelerating dl inference with (open)capi and posit numbers
04 accelerating dl inference with (open)capi and posit numbers04 accelerating dl inference with (open)capi and posit numbers
04 accelerating dl inference with (open)capi and posit numbers
 
Spil Storage Platform (Erlang) @ EUG-NL
Spil Storage Platform (Erlang) @ EUG-NLSpil Storage Platform (Erlang) @ EUG-NL
Spil Storage Platform (Erlang) @ EUG-NL
 
Crash course on data streaming (with examples using Apache Flink)
Crash course on data streaming (with examples using Apache Flink)Crash course on data streaming (with examples using Apache Flink)
Crash course on data streaming (with examples using Apache Flink)
 
Spark Summit EU talk by Berni Schiefer
Spark Summit EU talk by Berni SchieferSpark Summit EU talk by Berni Schiefer
Spark Summit EU talk by Berni Schiefer
 
Nafems15 Technical meeting on system modeling
Nafems15 Technical meeting on system modelingNafems15 Technical meeting on system modeling
Nafems15 Technical meeting on system modeling
 
A Lightweight Infrastructure for Graph Analytics
A Lightweight Infrastructure for Graph AnalyticsA Lightweight Infrastructure for Graph Analytics
A Lightweight Infrastructure for Graph Analytics
 
Hadoop performance optimization tips
Hadoop performance optimization tipsHadoop performance optimization tips
Hadoop performance optimization tips
 
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...
DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...
 
Top schools in gudgao
Top schools in gudgaoTop schools in gudgao
Top schools in gudgao
 
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
 
What is Distributed Computing, Why we use Apache Spark
What is Distributed Computing, Why we use Apache SparkWhat is Distributed Computing, Why we use Apache Spark
What is Distributed Computing, Why we use Apache Spark
 
Top schools in noida
Top schools in noidaTop schools in noida
Top schools in noida
 

More from openseesdays

Opensees integrated in a BIM workflow as calculation engine
Opensees integrated in a BIM workflow as calculation engineOpensees integrated in a BIM workflow as calculation engine
Opensees integrated in a BIM workflow as calculation engineopenseesdays
 
Recent advances in modeling soil-structure interaction problems using OpenSees
Recent advances in modeling soil-structure interaction problems using OpenSeesRecent advances in modeling soil-structure interaction problems using OpenSees
Recent advances in modeling soil-structure interaction problems using OpenSeesopenseesdays
 
Expert systems for advanced FE modelling of bridges and buildings using OpenSees
Expert systems for advanced FE modelling of bridges and buildings using OpenSeesExpert systems for advanced FE modelling of bridges and buildings using OpenSees
Expert systems for advanced FE modelling of bridges and buildings using OpenSeesopenseesdays
 
Implementation and finite-element analysis of shell elements confined by thro...
Implementation and finite-element analysis of shell elements confined by thro...Implementation and finite-element analysis of shell elements confined by thro...
Implementation and finite-element analysis of shell elements confined by thro...openseesdays
 
Development of an OpenSees model for collapse risk assessment of Italian-code...
Development of an OpenSees model for collapse risk assessment of Italian-code...Development of an OpenSees model for collapse risk assessment of Italian-code...
Development of an OpenSees model for collapse risk assessment of Italian-code...openseesdays
 
Blind test prediction of an infilled RC building with OpenSees
Blind test prediction of an infilled RC building with OpenSeesBlind test prediction of an infilled RC building with OpenSees
Blind test prediction of an infilled RC building with OpenSeesopenseesdays
 
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...openseesdays
 
A new Graphical User Interface for OpenSees
A new Graphical User Interface for OpenSeesA new Graphical User Interface for OpenSees
A new Graphical User Interface for OpenSeesopenseesdays
 
Assessment of the seismic performance of steel frames using OpenSees
Assessment of the seismic performance of steel frames using OpenSeesAssessment of the seismic performance of steel frames using OpenSees
Assessment of the seismic performance of steel frames using OpenSeesopenseesdays
 
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...openseesdays
 
Efficient analytical and hybrid simulations using OpenSees
Efficient analytical and hybrid simulations using OpenSeesEfficient analytical and hybrid simulations using OpenSees
Efficient analytical and hybrid simulations using OpenSeesopenseesdays
 
Numerical modelling of RC columns with plain reinforcing bars
Numerical modelling of RC columns with plain reinforcing barsNumerical modelling of RC columns with plain reinforcing bars
Numerical modelling of RC columns with plain reinforcing barsopenseesdays
 
OpenSees: Future Directions
OpenSees: Future DirectionsOpenSees: Future Directions
OpenSees: Future Directionsopenseesdays
 
OpenSees solver with a differential evolutionary algorithm for structural opt...
OpenSees solver with a differential evolutionary algorithm for structural opt...OpenSees solver with a differential evolutionary algorithm for structural opt...
OpenSees solver with a differential evolutionary algorithm for structural opt...openseesdays
 
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...An OpenSees material model for the cyclic behaviour of corroded steel bar in ...
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...openseesdays
 
Numerical investigation on the seismic behaviour of repaired and retrofitted ...
Numerical investigation on the seismic behaviour of repaired and retrofitted ...Numerical investigation on the seismic behaviour of repaired and retrofitted ...
Numerical investigation on the seismic behaviour of repaired and retrofitted ...openseesdays
 
Modelling with fibre beam elements for load capacity assessment of existing m...
Modelling with fibre beam elements for load capacity assessment of existing m...Modelling with fibre beam elements for load capacity assessment of existing m...
Modelling with fibre beam elements for load capacity assessment of existing m...openseesdays
 
Modelling of soil-structure interaction in OpenSees: A practical approach for...
Modelling of soil-structure interaction in OpenSees: A practical approach for...Modelling of soil-structure interaction in OpenSees: A practical approach for...
Modelling of soil-structure interaction in OpenSees: A practical approach for...openseesdays
 
Modelling of a shear reinforced flat slab building for seismic fragility anal...
Modelling of a shear reinforced flat slab building for seismic fragility anal...Modelling of a shear reinforced flat slab building for seismic fragility anal...
Modelling of a shear reinforced flat slab building for seismic fragility anal...openseesdays
 
Evaluating the use of OpenSees for lifetime seismic performance assessment of...
Evaluating the use of OpenSees for lifetime seismic performance assessment of...Evaluating the use of OpenSees for lifetime seismic performance assessment of...
Evaluating the use of OpenSees for lifetime seismic performance assessment of...openseesdays
 

More from openseesdays (20)

Opensees integrated in a BIM workflow as calculation engine
Opensees integrated in a BIM workflow as calculation engineOpensees integrated in a BIM workflow as calculation engine
Opensees integrated in a BIM workflow as calculation engine
 
Recent advances in modeling soil-structure interaction problems using OpenSees
Recent advances in modeling soil-structure interaction problems using OpenSeesRecent advances in modeling soil-structure interaction problems using OpenSees
Recent advances in modeling soil-structure interaction problems using OpenSees
 
Expert systems for advanced FE modelling of bridges and buildings using OpenSees
Expert systems for advanced FE modelling of bridges and buildings using OpenSeesExpert systems for advanced FE modelling of bridges and buildings using OpenSees
Expert systems for advanced FE modelling of bridges and buildings using OpenSees
 
Implementation and finite-element analysis of shell elements confined by thro...
Implementation and finite-element analysis of shell elements confined by thro...Implementation and finite-element analysis of shell elements confined by thro...
Implementation and finite-element analysis of shell elements confined by thro...
 
Development of an OpenSees model for collapse risk assessment of Italian-code...
Development of an OpenSees model for collapse risk assessment of Italian-code...Development of an OpenSees model for collapse risk assessment of Italian-code...
Development of an OpenSees model for collapse risk assessment of Italian-code...
 
Blind test prediction of an infilled RC building with OpenSees
Blind test prediction of an infilled RC building with OpenSeesBlind test prediction of an infilled RC building with OpenSees
Blind test prediction of an infilled RC building with OpenSees
 
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...
Modelling the out-of-plane behaviour of URM infills and the in-plane/out-of-p...
 
A new Graphical User Interface for OpenSees
A new Graphical User Interface for OpenSeesA new Graphical User Interface for OpenSees
A new Graphical User Interface for OpenSees
 
Assessment of the seismic performance of steel frames using OpenSees
Assessment of the seismic performance of steel frames using OpenSeesAssessment of the seismic performance of steel frames using OpenSees
Assessment of the seismic performance of steel frames using OpenSees
 
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...
Non-linear dynamic analyses of a 60’s RC building collapsed during L’Aquila 2...
 
Efficient analytical and hybrid simulations using OpenSees
Efficient analytical and hybrid simulations using OpenSeesEfficient analytical and hybrid simulations using OpenSees
Efficient analytical and hybrid simulations using OpenSees
 
Numerical modelling of RC columns with plain reinforcing bars
Numerical modelling of RC columns with plain reinforcing barsNumerical modelling of RC columns with plain reinforcing bars
Numerical modelling of RC columns with plain reinforcing bars
 
OpenSees: Future Directions
OpenSees: Future DirectionsOpenSees: Future Directions
OpenSees: Future Directions
 
OpenSees solver with a differential evolutionary algorithm for structural opt...
OpenSees solver with a differential evolutionary algorithm for structural opt...OpenSees solver with a differential evolutionary algorithm for structural opt...
OpenSees solver with a differential evolutionary algorithm for structural opt...
 
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...An OpenSees material model for the cyclic behaviour of corroded steel bar in ...
An OpenSees material model for the cyclic behaviour of corroded steel bar in ...
 
Numerical investigation on the seismic behaviour of repaired and retrofitted ...
Numerical investigation on the seismic behaviour of repaired and retrofitted ...Numerical investigation on the seismic behaviour of repaired and retrofitted ...
Numerical investigation on the seismic behaviour of repaired and retrofitted ...
 
Modelling with fibre beam elements for load capacity assessment of existing m...
Modelling with fibre beam elements for load capacity assessment of existing m...Modelling with fibre beam elements for load capacity assessment of existing m...
Modelling with fibre beam elements for load capacity assessment of existing m...
 
Modelling of soil-structure interaction in OpenSees: A practical approach for...
Modelling of soil-structure interaction in OpenSees: A practical approach for...Modelling of soil-structure interaction in OpenSees: A practical approach for...
Modelling of soil-structure interaction in OpenSees: A practical approach for...
 
Modelling of a shear reinforced flat slab building for seismic fragility anal...
Modelling of a shear reinforced flat slab building for seismic fragility anal...Modelling of a shear reinforced flat slab building for seismic fragility anal...
Modelling of a shear reinforced flat slab building for seismic fragility anal...
 
Evaluating the use of OpenSees for lifetime seismic performance assessment of...
Evaluating the use of OpenSees for lifetime seismic performance assessment of...Evaluating the use of OpenSees for lifetime seismic performance assessment of...
Evaluating the use of OpenSees for lifetime seismic performance assessment of...
 

Recently uploaded

Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...roncy bisnoi
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdfankushspencer015
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINESIVASHANKAR N
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxupamatechverse
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escortsranjana rawat
 
Extrusion Processes and Their Limitations
Extrusion Processes and Their LimitationsExtrusion Processes and Their Limitations
Extrusion Processes and Their Limitations120cr0395
 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Christo Ananth
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxupamatechverse
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Dr.Costas Sachpazis
 
Porous Ceramics seminar and technical writing
Porous Ceramics seminar and technical writingPorous Ceramics seminar and technical writing
Porous Ceramics seminar and technical writingrakeshbaidya232001
 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSISrknatarajan
 
Processing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxProcessing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxpranjaldaimarysona
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlysanyuktamishra911
 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...ranjana rawat
 
Glass Ceramics: Processing and Properties
Glass Ceramics: Processing and PropertiesGlass Ceramics: Processing and Properties
Glass Ceramics: Processing and PropertiesPrabhanshu Chaturvedi
 
UNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular ConduitsUNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular Conduitsrknatarajan
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...ranjana rawat
 

Recently uploaded (20)

Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdf
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptx
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
 
Extrusion Processes and Their Limitations
Extrusion Processes and Their LimitationsExtrusion Processes and Their Limitations
Extrusion Processes and Their Limitations
 
DJARUM4D - SLOT GACOR ONLINE | SLOT DEMO ONLINE
DJARUM4D - SLOT GACOR ONLINE | SLOT DEMO ONLINEDJARUM4D - SLOT GACOR ONLINE | SLOT DEMO ONLINE
DJARUM4D - SLOT GACOR ONLINE | SLOT DEMO ONLINE
 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptx
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
 
Porous Ceramics seminar and technical writing
Porous Ceramics seminar and technical writingPorous Ceramics seminar and technical writing
Porous Ceramics seminar and technical writing
 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSIS
 
Processing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxProcessing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptx
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
 
Glass Ceramics: Processing and Properties
Glass Ceramics: Processing and PropertiesGlass Ceramics: Processing and Properties
Glass Ceramics: Processing and Properties
 
Roadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and RoutesRoadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and Routes
 
UNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular ConduitsUNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular Conduits
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
 
Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024
 

A shared-filesystem-memory approach for running IDA in parallel over informal computer clusters

  • 1. National Technical University of Athens School of Civil Engineering A shared-filesystem-memory approach for running IDA in parallel over informal computer clusters D. Vamvatsikos National Technical University of Athens …and friends…. EOSD 2017 Opensees Days Europe Porto, June 19-20, 2017
  • 2. 2 Introduction • Seismic performance evaluation. How? • Incremental Dynamic Analysis is accurate but slow – Multiple nonlinear dynamic analyses, Multiple records – 1 CPU  Needs patience! • Can we run it faster? – Use N CPUs. – Parallelize each nonlinear analysis? OpenSeesMP – Parallelize IDA? Easier, repetitive, more efficient 
  • 3. 3 LA9 building • 9 stories, 1 basement, T1 = 2.3s • 2D model with internal gravity frame, P-Delta, beam-hinges
  • 4. 4 The IDA tracing problem • High variability  many records • Complex shapes  many runs • Focus: 20 recs, 12 runs each • One i5 core = 40 hours • Only for academic use!
  • 5. 5 How to run IDA in parallel? • Each record is completely independent from the others – Distribute single-record IDA tasks – Use up to 20 cores, pure linear efficiency • Within each records the runs are not independent – Distribute single runs  need new tracing algorithm – Up to 20x12=240 cores (ideally) – Dependencies  Runs may be wasted
  • 6. 6 Target Application Environment • MATLAB + OpenSees • Informal cluster of dissimilar processors – Multiple cores per physical processor – New and old PCs – Unreliable networks – Incompetent system admins – Random PC deaths (Win10 reboots, student magic….) • Need resilience over an unreliable “cluster”
  • 7. 7 Scheme 1: Distribute records • Shared memory approach – Everything stored in a file server (shared directory) – One job = One .mat file – Race conditions? Use directories as lockfiles • Master / slave model – Master assigns tasks (records) – Slaves run task, return one IDA curve per record – Master assigns tasks to self  Max efficiency Master CPU Slave CPUs Send record Receive IDA curve Assign record to self
  • 8. 8 Scheme 1: Coarse grained • Pros – Negligible communication overhead – Uses existing IDA tracing algorithms (hunt & fill) – Easy programming • Cons – Low scalability – Records / CPUs <> integer  CPUs are idling e.g. 20 records / 3 cpus  2 cpus run 7 records, one runs 6.
  • 9. 9 Scheme 2: Distribute records then runs • Master / slave-1 / slave-2 model – Master assigns tasks (records) – Slaves-1 assign single runs to slaves-2 – Master & slaves-1 assign runs to selves Master CPU Slave-1 CPUs Send record Receive IDA curve Assign record/runs to self Assign runs to self Slave-2 CPUs Send run IM Receive EDPs
  • 10. 10 Scheme 2: Medium-grained • Pros – Still low communication overhead – Excellent scalability – Almost linear efficiency – Minimal CPU idling • Cons – Tougher to program: Dynamic Allocation of Tasks – Slave-cores compete for jobs (first-come, first-serve) – Needs new IDA tracing algorithm
  • 11. 11 Serial Hunt & Fill IDA tracing • Hunt-up (large steps) • Bisect to find collapse • Fill the gaps going down • Hunt-up and Bisect are unpredictable due to collapse Hunt-up Fill-down Bisect
  • 12. 12 Parallel Hunt & Fill IDA tracing • 3 CPUs  more non- converged runs • Fill-in is still efficient • 14 runs instead of 12 • More CPUs in hunt-up or Bisect  more waste! • Still, more efficient that stepping algorithm Hunt-up Fill-down Bisect
  • 13. 13 Example: 20 records, 3 identical CPUs • Parallel hunt&fill achieves near linear performance (2.96/3 = 99%) • When the last 2 records were run, cpu3 was helping alternatively cpus 1,2
  • 14. 14 Performance Comparison • For 60 CPUs, each record uses 3+ cores simultaneously • Efficiency drops to 88% only
  • 15. How is this done in software? 15 • Master script: [anls]=runIDA_NDpx(anls,trace,fupdate,sharedmem,runmode); • Slave script [icase]=runIDA_NDpx_slave(sharedmem,masterflag,...,imultcpu); • One record subscript [icase]=runIDA_NDpx_slave_onerec(sharedmem,...,imultcpu); • One run subscript [icase]=runIDA_NDpx_slave_onerun(sharedmem,...,imultcpu); • Subscripts are run based on availability of jobs Analysis and tracing parameters Tcl file updating info for all runs & models Shared memory IP address 0 = partition records 1 = partition runs Assign a number to each core Set to 1 if master self-assigned the run
  • 16. 16 Application: Model parameter uncertainties • Monte-Carlo based – Use full IDA (Ibarra, Dolsek, Vamva & Frag) – Approximate IDA • Response Surface (Liel et al) • SPO2IDA (Frag & Vamva) + … • Moment-Estimation based – FOSM (Ibarra, Liel et al, Lee & Mosalam, Vamva & Frag) – PEM (Vamva & Frag) + …
  • 17. 17 Beam point-hinge model • Allows hardening & softening with pinching loops • Residual plateau terminates at ultimate rotation  True hinge • Means: ah = 10%, μc = 3, ac = -50%, r = 50%, μu = 6, aMy = 1
  • 18. 18 “Mean” model IDA curves • 30 records, 12 runs each • P-Delta and beam hinging  Clear flatlines • Summarization into fractiles • Med. collapse capacity • Large dispersion
  • 19. 19 200 median IDAs (real or approximate)
  • 20. 20 Let’s try to improve • Progressive application of LHS – Initial small sample – Double size by adding samples in each step – Stop when adequate accuracy • Allow sampling on a record-by-record basis – Old  full multi-record IDA for each parameter sample – New  single-record IDA for each sample – Recycle records if not enough – Not new ideas, e.g. Schotanus & Franchin
  • 21. 21 Progressive LHS (1) • Start with Npoints ≥ Nvars • Say 4 samples for 2 vars • Run analysis • Double the samples  8
  • 22. 22 Progressive LHS (1) • Start with Npoints ≥ Nvars • Say 4 samples for 2 vars • Run analysis • Double the samples  8
  • 23. 23 Progressive LHS (2) • Run analysis • Double the samples  16 • Double again  32 • Stop, e.g. when dispersion is “stable”
  • 24. 24 Good but not perfect • Pros – Excellent scalability – Removes problem of a priori definition of sample size – Handle large number of r.v’s: insignificants disappear! – Better coverage of sample space (like orthogonal LHS) • Cons – Cannot use fast Iman-Conover algorithm for correlation – Prefer genetic-algorithms, e.g. Charmpis&Panteli (2004) – These may be slower but more accurate
  • 25. 25 Include records in LHS sampling • If not enough records: Recycle + vary incident angle No X1 X2 … XN XN+1 XN+2 1 x1,1 x1,2 … x1,N ang1 Rec1 2 x2,1 x2,2 … x2,N ang2 Rec2 … … … … … … … M xM,1 xM,2 … xM,N angM RecM M+1 xM+1,1 xM+1,2 … xM+1,N angM+1 Rec1 M+2 xM+2,1 xM+2,2 … xM+2,N angM+2 Rec2 … … … … … … ….
  • 26. 26 Good but not perfect (again) • Pros – Handles large number of random vars ( > 500) – Can vary incident angle (More records is better though) – Place influential vars first to better capture correlation • Cons – Cannot use some “fast-IDA” techniques (e.g. Liel et al response surface, or Azarbakht-Dolsek priority lists, or SPO2IDA) – Cannot distinguish epistemic from aleatory (Do we care?)
  • 27. 27 Convergence • Stable after 4th generation • 160 samples • Stable after 5th generation •  320 samples • Note 270 random vars!
  • 28. 28 Compare with “mean” model • Medians differ! • Conservative bias due to correlation structure • Betas are similar • In contrast with other results • Still needs work!
  • 29. 29 Some concluding remarks • Is this worth it? – Why not OpenSeesSP / OpenSeesMP? – Why not HTCondor or similar? – Why Matlab and not Python? • Where to go now? – If found useful, easily ported to Python – Works for MSA, IDA, Cloud, any strategy – You can use this now: Will release by end of summer!