Modeling cloud failure data: a case study of the virtual computing lab

Authors:
Meiyappan Nagappan;Aaron Peeler;Mladen Vouk
Affiliations:
North Carolina State University, Raleigh, NC, USA;North Carolina State University, Raleigh, NC, USA;North Carolina State University, Raleigh, NC, USA
Venue:
Proceedings of the 2nd International Workshop on Software Engineering for Cloud Computing
Year:
2011

Citing 9
Cited 0

Handbook of software reliability engineering

Handbook of software reliability engineering
A logarithmic poisson execution time model for software reliability measurement

ICSE '84 Proceedings of the 7th international conference on Software engineering
Reliability and availability of a wide area network-based education system

ISSRE '96 Proceedings of the The Seventh International Symposium on Software Reliability Engineering
Quality of service modeling for wide area network based systems

Quality of service modeling for wide area network based systems
Market-Oriented Cloud Computing: Vision, Hype, and Reality for Delivering IT Services as Computing Utilities

HPCC '08 Proceedings of the 2008 10th IEEE International Conference on High Performance Computing and Communications
NCSU's Virtual Computing Lab: A Cloud Computing Solution

Computer
A view of cloud computing

Communications of the ACM
Characterizing cloud computing hardware reliability

Proceedings of the 1st ACM symposium on Cloud computing
Using VCL technology to implement distributed reconfigurable data centers and computational services for educational institutions

IBM Journal of Research and Development

Quantified Score

Hi-index	0.00

Visualization

Abstract

Virtual Computing Lab is a higher education cloud computing environment that on demand, allocates a chosen software stack on the required hardware and gives access to the customers, in this case NCSU students, faculty and staff. VCL has been in operation since 2004. An important component of the quality of the services provided by a cloud is the reliability and availability. For example, typical availability of the system exceeds 0.999, and reservation reliability is in the 0.99 range. VCL provides comprehensive information (provenance, logs, etc.) about its execution, its resources, and its performance. We mined the VCL log files to find out more about its reliability and availability, and the character of its faults and failures. This paper presents some of these results.