Presentation is loading. Please wait.

Presentation is loading. Please wait.

Communication layer * Agent message delivery filtering Framework layer * Spread agents on processors * Calling of the functions on agents in order * Agent.

Similar presentations


Presentation on theme: "Communication layer * Agent message delivery filtering Framework layer * Spread agents on processors * Calling of the functions on agents in order * Agent."— Presentation transcript:

1 Communication layer * Agent message delivery filtering Framework layer * Spread agents on processors * Calling of the functions on agents in order * Agent message transmission (MPI) * Input and output to files FLAME : A parallel agent based framework using X-machines Mariam Kiran Simon Coakley Mike Holcombe Department of Computer Science, University of Sheffield, Sheffield, UK Chris Greenough David Worth Lee Shawn Chin Rutherford Appleton Laboratory, UK X-machines are finite state machines with the inclusion of memory which influences the state transitions in the model. They have been used to specify and test software systems and are also being used for modelling more complex structures such as agents in agent based models. FLAME uses this paradigm accompanied with abilities to parallelize the models allowing high concentrations of agents with more complex structures to be simulated in finite time. Two examples from the fields of biology and economics have been described below as case studies.  Identify the system states and functions relevant to the system being modelled. This produces a state transition diagram.  Identify the inputs and outputs for each function. These could be the messages arriving or leaving influencing the functions.  For each state identify the memory variables being used.  Having identified the attributes each system function can be described as a separate X-machine resulting in a X-machine functional hierarchy. Biology: Keratinocyte cell modelEconomics: Labour market model We would like to acknowledge the works of Neil Walkinshaw and Phil McMinn in contributing to the modelling methodologies. Few Results Current works State transition diagrams for two agents - firm and household. State transition diagram for a cell showing the different forms it can exist in. Following from the input/output messages the function dependencies can be created. These allow how the different modules can be parallelised over a set of processors. FLAME is being currently being used in various projects belonging to different disciplines.  The Epitheliome Project is using the tool to model social behaviour of cells in epithelial tissues.  EURACE Project in an agent-based software platform for European economic policy design with heterogeneous interacting agents with new insights from a bottom up approach to economic modelling and simulation.  SUMO Systems Understanding of Microbial Oxygen Responses is studying the behaviour of the bacterium E- coli and its responses to the availability of oxygen. Various parallel machines are being used to test the optimal agent distribution of agents: FLAME’s Layer Structure Co-funded by the European Commission within the Sixth Framework Programme Model Layer * Define agents * Agent operations and their sequence * Set up the communication network Website: www.flame.ac.uk Introduction The dotted arrows represent data dependencies between the functions. These represent the synchronisation points which insure that all functions prior to that point have finished processing. These keeps track of all functions to be synchronised when running on multiple processors. 4 synchronisation points 3 synchronisation points Structure of a X- machine agent Acknowledgements  Mano – 1024 nodes of dual-core 700MHz PowerPC chips.  Hapu – 128 x 2.4GHz Opteron cores, with 2Gb memory per core.  NW_GRID – 32 SUN x 4100 nodes. Each node contains 2 Dual Core 2.4Ghz Opterons with 8GB of memory. That brings the total processor count to 192 Dual-Core Opterons (384 processor cores).  HPCx – Total of 2560 processors. Modelling


Download ppt "Communication layer * Agent message delivery filtering Framework layer * Spread agents on processors * Calling of the functions on agents in order * Agent."

Similar presentations


Ads by Google