Software—Practice & Experience
Evaluating architectures for multithreaded object request brokers
Communications of the ACM
Essential COM
Runtime Performance Modeling and Measurement of Adaptive Distributed Object Applications
On the Move to Meaningful Internet Systems, 2002 - DOA/CoopIS/ODBASE 2002 Confederated International Conferences DOA, CoopIS and ODBASE 2002
JaViz: a client/server Java profiling tool
IBM Systems Journal
Call path profiling of monotonic program resources in UNIX
Usenix-stc'93 Proceedings of the USENIX Summer 1993 Technical Conference on Summer technical conference - Volume 1
Debugging distributed object applications with the Eclipse platform
eclipse '04 Proceedings of the 2004 OOPSLA workshop on eclipse technology eXchange
Debugging and testing middleware with aspect-based control-flow and causal patterns
Proceedings of the 9th ACM/IFIP/USENIX International Conference on Middleware
An eclipse-based tool for symbolic debugging of distributed object systems
OTM'07 Proceedings of the 2007 OTM Confederated international conference on On the move to meaningful internet systems: CoopIS, DOA, ODBASE, GADA, and IS - Volume Part I
Detecting and Coordinating Complex Patterns of Distributed Events with KETAL
Electronic Notes in Theoretical Computer Science (ENTCS)
Hi-index | 0.00 |
Current software development techniques and toolslack the capability to characterize function call chains inmultithreaded and distributed applications built uponcomponent technologies like CORBA, COM and J2EE.The root cause is that causal linkage informationnecessary to trace end-to-end call chains is private toeach vendor's runtime and often unavailable for loggingor analysis. We propose and demonstrate a mechanismfor maintaining and correlating global causalityinformation of component-based applications, and usingthis information to expose and characterize function callchains and their associated behaviors in suchmultithreaded and distributed applications. Our approachrelies on a global virtual tunnel facilitated by theinstrumented stubs and skeletons. This tunnel maintainsand correlates causal information throughout the end-to-endcall chains spanning threads, processes andprocessors. As a result, monitoring data captured locallycan be correlated and system-wide propagation of timinglatency and CPU utilization becomes perceivable.