PIM

The PIM (Process Integrated Mechanism)

The PIM is a novel coordination approach proposed by Dr. Kenneth M. Ford and Prof. Niranjan Suri at the Institute for Human and Machine Cognition (IHMC), Pensacola, Florida, USA.

The core idea of the PIM approach is to retain the perspective of the single controlling authority but abandon the notion that it must have a fixed location within the system. Instead, the coordinating process is rapidly moved among the component parts of the PIM. From this perspective, the operation of the PIM is the inverse of multiprocessing or timesharing. Instead of rapidly switching a single CPU between multiple processes, the PIM rapidly switches a single coordinating process between multiple CPUs in multiple physical entities. Just as a multitasking system effectively hides the process switching, the PIM runtime system effectively makes process cycling transparent.<ref>K.M.Ford, N.Suri, K.Kosnar, P.Jisl, P.Benda, M.Pechoucek, L.Preucil, "A Game-based Approach to Comparing Different Coordination Mechanisms" DHMS2008</ref>

Three Institutions are working on different pieces of the PIM project:

• IHMC
• University of Modena and Reggio Emilia
• Czech Technical University in Prague

The Unimore Group

The group at the University of Modena and Reggio Emilia is composed by:

• Prof. Giacomo Cabri
• Prof. Letizia Leonardi
• PhD Raffaele Quitadamo
• Ing. Danilo Ansaloni (Research Associate at IHMC)

The PIM architecture

The issue of strong mobility of Java threads

• Mobile JikesRVM implementation

First proposed model

Here you can find brief explanations about the first proposed model we developed:

• General model discussions

• Programming the CP for a Better Reactivity

• Coping with Fault-tolerance

• The caching mechanism

• Using the GroupManager on top of Mobile JikesRVM

A proposal for the new PIM model

Here we discuss about the new PIM model on which we are currently working on:

• The new PIM architecture
• Reducing network traffic and power consumption
• Dealing with partially connected network topologies
• The neglet tolerance
• The fault tolerance mechanism
• The caching mechanism
• Tradeoffs
• Pseudo-code

Learning from the past

• A comparison between the PIM model and other multirobot coordination approaches
• Possible critics to the PIM model

Old stuff

Old Stuff

Footnotes

<references />