TY - GEN
T1 - Survivability of a distributed multi-agent application - A performance control perspective
AU - Gnanasambandam, Nathan
AU - Lee, Seokcheon
AU - Kumara, Soundar R.T.
AU - Gautam, Natarajan
AU - Peng, Wilbur
AU - Manikonda, Vikram
AU - Brinn, Marshall
AU - Greaves, Mark
N1 - Funding Information:
Epidemic algorithms can be used for managing replicated data [4-9]. In an epidemic approach, sites perform update operations locally and communicate peer-to-peer in a lazy manner to propagate updates. Transactional consistency is achieved by * Supported by the Natural Science Fundation of Zhejiang Province, China (Grant no. M603230) and the Research Fund for Doctoral Program of Higher Education from Ministry of Education, China (Grant no. 20020335020).
Funding Information:
Supported by the Natural Science Fundation of Zhejiang Province, China (Grant no. M603230) and the Research Fund for Doctoral Program of Higher Education from Ministry of Education, China (Grant no. 20020335020).
PY - 2005
Y1 - 2005
N2 - Distributed Multi-Agent Systems (DMAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. The utility from maintaining performance is an important component of their survivability. This utility is often met by identifying trade-offs between quality of service and performance. To adoptively choose the operational settings for better utility, we propose an autonomous and scalable queueing theory based methodology to control the performance of a hierarchical network of distributed agents. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone. When the problem scales, another larger queueing network could be composed using zones as bu0ilding-blocks. This method advocates the systematic specification of the DMAS's attributes to aid real-time translation of the DMAS into a queueing network. We prototype our framework in Cougaar and verify our results.
AB - Distributed Multi-Agent Systems (DMAS) such as supply chains functioning in highly dynamic environments need to achieve maximum overall utility during operation. The utility from maintaining performance is an important component of their survivability. This utility is often met by identifying trade-offs between quality of service and performance. To adoptively choose the operational settings for better utility, we propose an autonomous and scalable queueing theory based methodology to control the performance of a hierarchical network of distributed agents. By formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance and subsequently the utility, from which we identify the control alternative for a localized, multi-tier zone. When the problem scales, another larger queueing network could be composed using zones as bu0ilding-blocks. This method advocates the systematic specification of the DMAS's attributes to aid real-time translation of the DMAS into a queueing network. We prototype our framework in Cougaar and verify our results.
UR - http://www.scopus.com/inward/record.url?scp=33745700184&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745700184&partnerID=8YFLogxK
U2 - 10.1109/MASSUR.2005.1507044
DO - 10.1109/MASSUR.2005.1507044
M3 - Conference contribution
AN - SCOPUS:33745700184
SN - 078039447X
SN - 9780780394476
T3 - 2005 IEEE 2nd Symposium on Multi-Agent Security and Survivability
SP - 21
EP - 30
BT - 2005 IEEE 2nd Symposium on Multi-Agent Security and Survivability
T2 - 2005 IEEE 2nd Symposium on Multi-Agent Security and Survivability
Y2 - 30 August 2005 through 31 August 2005
ER -