Abstract
Popular public cloud infrastructures tend to feature centralized, mutual exclusion models for distributed resources, such as file systems. The result of using such centralized solutions in the Google File System (GFS), for instance, reduces scalability, increases latency, creates a single point of failure, and tightly couples applications with the underlying services. In addition to these quality-of-service (QoS) and design problems, the GFS methodology does not support generic priority preference or pay-differentiated services for cloud applications, which public cloud providers may require under heavy loads. This paper presents a distributed mutual exclusion algorithm called Prioritizable Adaptive Distributed Mutual Exclusion (PADME) that we designed to meet the need for differentiated services between applications for file systems and other shared resources in a public cloud. We analyze the fault tolerance and performance of PADME and show how it helps cloud infrastructure providers expose differentiated, reliable services that scale. Results of experiments with a prototype of PADME indicate that it supports service differentiation by providing priority preference to cloud applications, while also ensuring high throughput.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
