Informatics 2010
DOI: 10.2316/p.2010.724-031
|View full text |Cite
|
Sign up to set email alerts
|

Using Virtualization to Validate Fault-Tolerant Distributed Systems

Abstract: Asynchronous events and complexs ystem state distributed across independent nodes makee xposure and diagnosis of flaws in distributed systems a challenge. The difficulties are exacerbated when the goal is to validate fault tolerance mechanisms that are activated only by the occurrence of errors, which are, by nature, rare. Va lidation of fault tolerance mechanisms is often done by injecting faults that emulate the actual faults and ''stress''t he functionality of the resilience mechanisms. Va lidation campaign… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
4
0

Year Published

2011
2011
2021
2021

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(4 citation statements)
references
References 16 publications
0
4
0
Order By: Relevance
“…The goal is to first expose problem areas and then evaluate the effectiveness of the refined mechanisms to recover from a variety of system corruptions. We used the UCLA Gigan fault injector [16,9,22] to inject faults into different VI components. Gigan can reside in the VMM and inject many types of faults into the VMs and the VMM.…”
Section: Fault Injection Campaigns and Fault Typesmentioning
confidence: 99%
“…The goal is to first expose problem areas and then evaluate the effectiveness of the refined mechanisms to recover from a variety of system corruptions. We used the UCLA Gigan fault injector [16,9,22] to inject faults into different VI components. Gigan can reside in the VMM and inject many types of faults into the VMs and the VMM.…”
Section: Fault Injection Campaigns and Fault Typesmentioning
confidence: 99%
“…To decrease the time of restoring pristine disk images for each fault injection run, a stackable union file system (AUFS [25]) is used [11]. Each AppVM creates twoR AID-1 devices, one containing the pristine disk state as a read-only branch and the second containing a smaller empty write branch.…”
Section: Fault Injection Campaignmentioning
confidence: 99%
“…We use the UCLA Gigan fault injector [14,11] to inject single bit flip faults into registers (general purpose and the program counter) while CPUs are executing VMM and DVM (user and kernel level) code. While these injected faults do not accurately represent all possible faults, theya re a good choice since transient hardware faults in CPU logic and memory are likely to be manifested as erroneous values in registers.…”
Section: Fault Injection Campaignmentioning
confidence: 99%
“…We use the UCLA Gigan fault injector [12,9] to inject single bit flip faults into CPU registers (general purpose and the program counter) while the CPUs are executing VI code. This type of injection is used since it causes arbitrary corruptions in the VI components.…”
Section: Methodsmentioning
confidence: 99%