Research of dynamic load balancing based on the UMTS-SGSN system

Furong Wang; Zuwei Teng

doi:10.1117/12.523402

28 April 2004 Research of dynamic load balancing based on the UMTS-SGSN system

Furong Wang, Zuwei Teng

Proceedings Volume 5284, Wireless Communications and Networks; (2004) https://doi.org/10.1117/12.523402
Event: Asia-Pacific Optical and Wireless Communications, 2003, Wuhan, China

Abstract

During the distributed design of UMTS-SGSN system, the Dynamic Central Tasks Scheduling algorithm (DCTS) with good real-time property is researched to meet the hard requirements of bandwidth and latency of UMTS-SGSN system. Based on the assumption without task migration, Markov queuing network model is used to model UMTS-SGSN system instead of M/M/c queuing model. For single user with one PDP context, DCTS is optimal through the theoretic calculation and simulation. And for single user with not less than one PDP context, the performance of DCTS is discussed through the simulation. In the case of longer service time in data service, DCTS is compared with stochastic scheduling algorithm and cycle scheduling algorithm. Simulation results show that DCTS is better than other two algorithms, and when service time is shorter, three algorithms' performance is approximately same. The performances of the three algorithms in different service time are also analyzed by adjusting the parameter μ of the exponential distribution. Results show that exponential distribution's heavy-tailed effect of the service time leads to deteriorating the efficiency of load balancing and DCTS shall initiate less task migration and decrease system overhead.

Citation Download Citation

Furong Wang and Zuwei Teng "Research of dynamic load balancing based on the UMTS-SGSN system", Proc. SPIE 5284, Wireless Communications and Networks, (28 April 2004); https://doi.org/10.1117/12.523402

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available