Abstract | ||
---|---|---|
The migration of mission-critical workloads to the cloud and the automation of various aspects of datacenter management is contributing to the evolution of software-defined infrastructures. One implication of this evolution is that the composition (both physical and virtual) and logical topology of datacenters is becoming even more dynamic. Identification of performance problems (e.g. bottlenecks) in such environments needs to be done with awareness of this dynamic topology to understand the impact of dependencies among components. A technique is introduced that a) employs expert knowledge to identify bottleneck components using associated performance metrics, and b) utilizes dynamic dependencies to rank problem components in order to facilitate diagnosis efforts. The technique is demonstrated experimentally on an OpenStack testbed with realistic fault injection. Results of experiment case studies show that the technique is able to correctly detect and rank problem nodes. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/ACIT-CSII-BCD.2017.40 | 2017 5th Intl Conf on Applied Computing and Information Technology/4th Intl Conf on Computational Science/Intelligence and Applied Informatics/2nd Intl Conf on Big Data, Cloud Computing, Data Science (ACIT-CSII-BCD) |
Keywords | DocType | ISBN |
Cloud Performance Monitoring,Analysis and Diagnosis,Performance Anomaly Detection,Cloud Resource Management,Software-defined Infrastructures | Conference | 978-1-5386-3303-8 |
Citations | PageRank | References |
0 | 0.34 | 11 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Olumuyiwa Ibidunmoye | 1 | 38 | 2.66 |
Thijs Metsch | 2 | 0 | 0.34 |
Victor Bayon-Molino | 3 | 1 | 0.77 |
Erik Elmroth | 4 | 1675 | 149.84 |