Performance Issues in TACOMA

• Main Authors: Dag Johansen, Nils P. Sudmann, Robbert Van Renesse
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Published: 1997
• Subjects: Tromsø
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.3630; http://www.tacoma.cs.uit.no/papers/ECOOP.tacoma.ps

. Mobile code performance depends, in part, on the costs of transferring an agent from one host to another and of initiating execution of that agent on a target host. These costs are reported for TACOMA (Tromsø and COrnell Moving Agents) v1.3, a UNIX-based system that supports agents. The experiments suggest opportunities for performance enhancements, both by changing the underlying operating system and by changing the architecture of the TACOMA run-time system. 1 Introduction One of the motivations for using mobile code in distributed applications is the potential for improved performance. Moving a program between hosts in a network may be cheaper than moving large amounts of data between those hosts. Of course, the performance improvement will depend on the relative cost of moving and installing code. This paper describes detailed measurements of move and install operations for mobile code that is run using the TACOMA (Tromsø and COrnell Moving Agents) system 3 . In TACOMA, a piec.