Presentation 44,45
Presentation 43
Tactical C4I Grid Implementation Challenges
Adi Sussholz Vice President - Chief Technology Officer
Elbit Systems Ltd
*******************
Elbit Systems Ltd. is an international defense electronics company engaged in a wide range of defense related programs throughout the world, in the areas of aerospace, land and naval systems command, control, communications, computers, intelligence, surveillance and reconnaissance ( C41SR ), advanced electro-optic and space technologies.
In 2002 Elbit was selected to be the prime contractor of the Digital Army Program - DAP. This program brought the NCW (Network Centric Warfare) revolution to the Israeli army on all levels from the single platform up to the regional command. Since end 2004 the program is operational. The hart of the DAP program is the Tactical C41 Grid.
Both the IDF C2 operational methodology and the NCW are distributed in essence. The tactical military environment is highly unstable and dynamic in nature. The Tactical C41 Grid must support these operational properties; each node must behave according to its surrounding environment (including other nodes). Such a system exhibits non linear behavior - small changes in the node's behavior might cause big changes in the system's behavior.
The presentation will describe the challenges in implementing our Tactical C41 grid and the methodology used to do so.
******************************
Presentation 44
The MOSIX Organizational Grid System
Prof. Amnon Barak and Amnon Shiloh
Department of Computer Science, The Hebrew University of Jerusalem
******************************
MOSIX is virtualization software that extends Linux with capabilities for adaptive management of computational resources in a cluster or in an organizational Grid. Its main feature is to make all the participating computers perform like a single computer with multiple processors, almost like an SMP. Users can run parallel and sequential applications by creating multiple processes, then let MOSIX seek resources and automatically migrate processes among nodes to improve the overallperformance, without changing the run-time environment of the migrated processes.
Originally developed to manage a single cluster, e.g., in a department, MOSIX was recently extended with new features for management of an organizational (trusted) Grid, with several clusters, e.g.,in different departments. Its main new features are: Uniform management within a cluster and across the Grid, including automatic resource discovery and a flexible partitioning of nodes to private virtual clusters.Migrated processes run in a secure run time environment (sandbox) which prevents access to local resources, e.g. files, in the hosting nodes.
Support of disruptive configurations, e.g., clusters can join or leave the Grid at any time. Multiple, evacuated processes are frozen rather than killed. Flood control prevents overloading of nodes. Support of batch jobs, checkpoint and recovery and live queuing - queued jobs preserve their full generic Linux environment. An on-line monitor for the status of the Grid resources and each cluster.
For over 10 years, large production MOSIX clusters are used to run scientific applications, e.g., genomic, molecular dynamics, nano-technology; engineering applications, e.g., CFD, weather forecasting, crash simulations, oil exploration, VLSI design, pharmaceutical; financial modeling, rendering, compilation farms and defense related applications. Further information,including a demo of a 13 cluster MOSIX Grid with over 300 nodes, is available at http://www.mosix.org
