1 / 26

MASS: A Parallel platform for Agent -Based Computing

MASS: A Parallel platform for Agent -Based Computing. Munehiro Fukuda, Ph.D. Computing & Software Systems University of Washington Bothell. Outline. Software agents Research Area 1: Agent-Based Models Research Area 2: Agent-Based Big Data Analysis Student Projects

sidneym
Download Presentation

MASS: A Parallel platform for Agent -Based Computing

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CSS599 Faculty Research MASS: A Parallel platform for Agent-Based Computing Munehiro Fukuda, Ph.D. Computing & Software Systems University of Washington Bothell

  2. CSS599 Faculty Research Outline • Softwareagents • Research Area 1: Agent-Based Models • Research Area 2: Agent-Based Big Data Analysis • Student Projects • Past students achievements • Currentand future student projects • Expected activities

  3. CSS599 Faculty Research Software Agents Cognitive Agents ReactiveAgents fine-grain entities, each reacting to its environment with simple rules. • coarse-grain execution entities that achieve network-administrative and computation-intensive task, based on their behavioral intelligence Ourfocus

  4. CSS599 Faculty Research Our Research Areas Agent-BasedModeling Agent-Based Data Sciences Analyzing distributed data by dispatching agents Example: Network problems such as triangle counting, BFS-based shortest path Biologically inspired optimization • Simulating an emergent collective group behavior of agents • Example: • Traffic simulation • Epidemic simulation

  5. CSS599 Faculty Research ResearchArea 1Agent-BasedSimulation

  6. CSS599 Faculty Research Conventional Traffic SimulationMacroscopic Simulation • Merits • Mathematical models/verification, • Wave differential equation • Use of WSDOT annual traffic report • Demerits • No micro events or interaction considered: • Traffic singnals and lanes • Parking • No dynamic events considered: • Weather • Dynamic trip plans Uncertain benefits: Application of Bayesian melding to the Alaskan Way Viaduct in Seattle

  7. CSS599 Faculty Research Agent-BasedTrafficSimulation • Micro-simulation • Model each traveler as an agent • Consider as many traffic events as possible • Simulate traffic as an interaction among travelers and events • Merits • More realistic • Parallelization of agents rather than formulae • Demerits • Calibration required • Scalable computation required Event-basedqueuingsimulation Log File Score Statistics Leg Travel Distance Statistics Events Trip Durations Optimization is performed in terms of agents’ plans. 10% agents: reroute their plansdynamically. 90% agents: choose their best score.

  8. CSS599 Faculty Research Multi-Agent Spatial Simulation

  9. func1( ) func1( ) func1( ) func1( ) func2( ) func1( ) … … func2( ) func2( ) func3( ) func2( ) func3( ) CSS599 Faculty Research MASS Specification Public static void main( String[ ] args ) { MASS.init( args ); Places space = new Places( handle, “MySpace”, params, xSize, ySize); Agents agents = new Agents( handle, “MyAgents”, params, space, population ); space.callAll( MySpace.func1, params ); space.exchangeAll( MySpace.func2, neighbors ); agents.exchangeAll( MyAgents.func3 ); agents.manageAll( ); MASS.finish( ); }

  10. CSS599 Faculty Research Implementation Status • The latest status is given at: • http://depts.washington.edu/dslab/MASS/index.html • MASS Java • Intended for agent-based big data. • MASS C++and MASS GPU • Intended for agent-based models. • Debugger • Available for MASS Java and C++ versions • Availability • Pilot used in CSS534 since summer 2012 • Java and C++ versions arealdy released to publicly • https://bitbucket.org/mass_library_developers/ • https://bitbucket.org/mass_application_developers/

  11. CSS599 Faculty Research Static Agents versus Dynamic Agents Bank Bail In/Out TB Game of Life VDT Trasport Social Network Artificial Network

  12. CSS599 Faculty Research Programmability and Performance Programmability: Lines of Code Less parallel programming construct More code for agent descriptions Execution performance: msec over 8 computing nodes.

  13. CSS599 Faculty Research ResearchArea 2Agent-BasedDataSciences

  14. CSS599 Faculty Research Conventional v.s. Agent-Based Approach Data Streaming Data Hunting by Agents inputs RDD1 Map Transformation Big Data Temp RDD1 Reduce Action RDD1 results

  15. CSS599 Faculty Research Triangle Counting MapReduce, Spark, and Storm Agent-based Big Data Analytics actively move over data. Only necessary portionofdatacanbeexamined. Data can maintain its structure. Code is written from the driver’s viewpoint. • Analytics stay passive to receive data. • All data must be examined. • Data must be flattend to feed. • (1,2), (2,1), (2,3), (3,2), (2,4), (4,2), (2,5), (5,2), (3,4), (4,3), (4,5), (5,4) 1 • (1, [2]), (2, [1,3,4,5]), (3, [2,4]), (4, [2,3,5]), (5, [2,4]) • [(1,2), -], [(1,3), 2], [(2,3), -], [(2,3), 4], [(2,4), -], [(2,4),3], • [(2,4), 5], [(2,5), 4], … 2 5 • (2,3) [-,4]  (2,3,4), (2,4) [-,3,5] -> (2,4,3), (2,4,5)… • (2,3,4), (2,3,4), (2,3,4), (2,4,5), (2,4,5).. • (2,3,4), (2,4,5) 3 4

  16. CSS599 Faculty Research Potential Applications BFS Triangle Counting Ant Colonial Optimization Best Tool for K Means KNN Particle Swarm Optimization

  17. CSS599 Faculty Research Programmability and Performance

  18. CSS599 Faculty Research Past StudentResearchOutcome(gradudate students in red and undergraduate students in blue, listed below) • Yun-Ming Shih, Collin Gordon, Munehiro Fukuda, Jasper van de Ven, Christian Frekas, Translation of String-and-Pin-Based Shortest Path Construction into Data-Scalable Agent-Based Computational Models, WSC 2018, Gothemburg, Sweden, December 10-13 2018 • Craig Shih, Caleb Yang, Munehiro Fukuda, Benchmarking the Agent Descriptivity of Multi-Agent Simulators, In Proc. of PAAMS/MAS&S, Toledo, Spain, June 20-22, 2018 • Collin Gordon, Munehiro Fukuda, Demonstration of Agent-Based Clustering and Classification Algorithms, In Proc. of PAAMS/MAS&S, Toledo, Spain, June 20-22, 2018 • Christopher Bowzer, Benjamin Phan, Kasey Cohen, Munehrio Fukuda, Collision-Free Agent Migration in Spatial Simulation, In Proc. of the Joint Agent-oriented Workshops in Synergy JAWS, Prague, Czech Republic, September 3-6, 2017 • Jason Woodring, Matthew Sell, Munehiro Fukuda, Hazeline Asuncion, Eric Salathe, A Multi-Agent Parallel Approach to Analyzing Large Climate Data Sets, The 37th IEEE International Conference on Distributed Computing Systems, pages 1639-1648, Atlanta, GA, June 5-8, 2017 • Andrew Andersen, Wooyoung Kim, and Munehiro Fukuda, MASS-based NemoProfile Construction for an Efficient Network Motif Search, 2016 Big Data and Cloud Computing – BDCloud’16, October 8-10, 2016 • Zhiyuan Ma, Munehiro Fukuda, A Multi-Agent Spatial Simulation Library for Parallelizing Transport Simulation, the 2015 Winter Simulation Conference - WSC 2015, Newport Beach, December 6-9, 2015 –Received another paper invitation • BhargavMistry, Munehiro Fukuda, Dynamic Load Balancing in Multi-Agent Spatial Simulation, submitted to 2015 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing - PACRIM’15, Victoria, BC, Canada, August 24-26, 2015 –Best Paper Award • Matthew Kipps, Wooyoung Kim, Munehiro Fukuda, Agent and Spatial Based Parallelization of Biological Network Motif Search, submitted to 17th IEEE Int’l Conference on High Performance and Communication, New York, August 24-26, 2015 • Brett Yasutake, Niko Simonson, Jason Woodring, Nathan Duncan, William Pfeffer, Hazeline Asuncion, Munehiro Fukuda, Eric Salathe, Supporting Provenance in Climate Science Research, In Proc. of 7th International Conference on Information, Process, and Knowledge Management, eKnow 2015, Lisbon, Portugal, February 22-27, 2015 –Best Paper Award • Timothy Chuang, Munehiro Fukuda, A Parallel Multi-Agent Spatial Simulation Environment for Cluster Systems, In Proc. of the 16th IEEE International Conference on Computational Science and Engineering - CSE 2013, pages 143-150, Sydney, Australia, December 3-5, 2013 • SomuJabayalan, Munehiro Fukuda, Field-Based Job Dispatch and Migration, In Proc. of 2013 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing - PACRIM'13, pages 52-57, Victoria, BC, Canada, August 27-29, 2013

  19. CSS599 Faculty Research MASS Project Summary(gradudate students in red and undergraduate students in blue, listed below)

  20. CSS599 Faculty Research BenchmarkingandApplications Done in black and in-progress or TBD in red: • MASSCUDA: Computation demands • Toy Applications: Heat2D,Sugarscape • Benchmark: 7SBC(Social, Behavioral,andEconomic) applications • MASSC++: Computation/Spatial demands • Benchmark: 7SBC(Social, Behavioral,andEconomic) applications • Practical Applications: Influenza epidmic to redo or brand-new applications • MASSJava: Spatial or interactive demands • Benchmark A: 6 data science / graph applications • Benchmark B: Machine learning applications • Practical Applications: extension of climate analysis, network motif, etc. or brand-new applications

  21. CSS599 Faculty Research Hybrid, Multi GPUs, GPU Cluster main Entire simulation space Host memory Hybrid Cache and Run Multi GPUs Device memory GPU threads GPU threads Cluster

  22. CSS599 Faculty Research Event-Driven/Discrete-EventComputation Departure( ) onArrival( ) Next vertex Current vertex Previous vertex Event Driven Discrete Events Moving agents at each discrete logical time (= link weight) @onArrival Object footmark( ) { …; } @onDeparture( ) Object diffuse( ) { …; } 7 2 3 6 5 4 2 T = 0 T = 6 T = 11 T = 2

  23. CSS599 Faculty Research GraphGenerationandMapping CSV File Src, Dst1, Dst2, Dst3, Dst4, … 0, 5, 2, 3 1, 2, 3 3, 4

  24. CSS599 Faculty Research InteractiveComputation • CreateahistoryofnewPlaces,newAgents,callAll,exchangeAll,andmanageAllinvocations. • Rerun a history to a given checkpoint whenever a user wants to toss up the latest computation and resume old results. p = new Places( ); a = new Agents( ); a.callAll( ); a.manageAll( ); p = new Places( ); a = new Agents( ); a.callAll( ); p = new Places( ); a = new Agents( ); p = new Place( ); Rollback & Retrieve

  25. CSS599 Faculty Research Expected Activities • Research Group Meetings • Weekly status report to your lab mates • Quarterly research presentation • OptionalIndividual Meetings • Weekly progress discussions, code reviews, etc. with theprofessor, (i.e.,me). • Programming • We are a product-oriented group. You need to code something. • Publications • I recommend you to coauthor conference/journal papers.

  26. CSS599 Faculty Research Questions? • Please check: http://depts.washington.edu/dslab/ 2017 2018

More Related