Cube Data Model for Multilevel Statistics Computation of Live Execution Traces

 

In this paper, we proposed an efficient architecture, the corresponding data structures and algorithms to compute multilevel, multidimensional statistics of different system metrics over a stream of execution trace events.

The details of the proposed method and experimental results are presented in the paper.

Date: 
Thursday, April 18, 2013
Publication authors (members): 
Michel Dagenais
Naser Ezzati Jivan
Publisher: 
Wiley Concurrency and Computation: Practice and Experience
Publication version number: 
1.00
Bibtex: 
@article {CPE:CPE3272, author = {Ezzati-Jivan, Naser and Dagenais, Michel R.}, title = {Cube data model for multilevel statistics computation of live execution traces}, journal = {Concurrency and Computation: Practice and Experience}, volume = {27}, number = {5}, issn = {1532-0634}, url = {http://dx.doi.org/10.1002/cpe.3272}, doi = {10.1002/cpe.3272}, pages = {1069--1091}, keywords = {stream processing, multilevel analysis, OLAP analysis, trace abstraction, Linux kernel}, year = {2015}, }
Publication status: 
Submitted

Tracks concerned: