Visualization of Distributed Program Execution

• Published in: 2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT) pp. 1 - 6
• Main Authors: Satwik, Bhamidipati, Manoj, Kalakoti Girija, Immadi, Sai Rajendra, Jayaraman, Swaminathan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.07.2021
• Subjects: causal order; Data visualization; Debugging; Distributed databases; distributed systems; dynamic analysis; execution traces; Java; Message passing; partial order; Prototypes; sequence diagrams; Systematics; visualization
• DOI: 10.1109/ICCCNT51525.2021.9579575

Comprehending distributed programs remain a challenge despite the availability of sophisticated IDEs and debuggers. This is mainly due to the inability to get an integrated view of the program execution since distributed programs run as individual fragments on different systems. Our work integrates the fragments into a single meaningful whole and thereby aims to generate a paradigm that complements the user's way of debugging distributed systems. The context of work is the Java Interactive Visualization Environment, a state-of-the-art debugging, analysis, and visualization tool for Java programs. We elucidate a systematic approach to combine the execution traces of the individual fragments by preserving the causal order of message passing events. This forms the basis for the unified sequence diagram which provides the history of the execution and thus depicting the control view. We present the detailed steps of the construction. The prototype version of the integrated data view is achieved by aggregating the state vectors from individual fragments and tracing the transitions in order to produce a unified state diagram. Further work is in progress to derive a systematic approach to construct distributed state diagram.