Tracking data structures for postmortem analysis (NIER track)

Analyzing the runtime behaviors of the data structures is important because they usually relate to the obscured program performance and understanding issues. The runtime evolution history of data structures creates the possibility of building a lightweight and non-checkpointing based solution for th...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiao, Xiao, Zhou, Jinguo, Zhang, Charles
Format: Conference Proceeding
Language:English
Subjects:
Arrays
data structure
Debugging
History
Instruments
Monitoring
program analysis
Runtime
Software and its engineering > Software creation and management > Software verification and validation > Process validation > Traceability
Software and its engineering > Software creation and management > Software verification and validation > Software defect analysis > Software testing and debugging
tracing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Analyzing the runtime behaviors of the data structures is important because they usually relate to the obscured program performance and understanding issues. The runtime evolution history of data structures creates the possibility of building a lightweight and non-checkpointing based solution for the backward analysis for validating and mining both the temporal and stationary properties of the data structure. We design and implement TAEDS, a framework that focuses on gathering the data evolution history of a program at the runtime and provides a virtual machine for programmers to examine the behavior of data structures back in time. We show that our approach facilitates many programming tasks such as diagnosing memory problems and improving the design of the data structures themselves.
ISSN:0270-5257
1558-1225
DOI:10.1145/1985793.1985938