Abstract
This paper proposes a new resource management scheme that supports SLA (Service-Level Agreement) in a bigdata distributed storage system. Basically, it makes use of two mapping modes, isolated mode and shared mode, in an adaptive manner. In specific, to ensure different QoS (Quality of Service) requirements among clients, it isolates storage devices so that urgent clients are not interfered by normal clients. When there is no urgent client, it switches to the shared mode so that normal clients can access all storage devices, thus achieving full performance. To provide this adaptability effectively, it devises two techniques, called logical cluster and normal inclusion. In addition, this paper explores how to exploit heterogeneous storage devices, HDDs (Hard Disk Drives) and SSDs (Solid State Drives), to support SLA. It examines two use cases and observes that separating data and metadata into different devices gives a positive impact on the performance per cost ratio. Real implementation-based evaluation results show that this proposal can satisfy the requirements of diverse clients and can provide better performance compared with a fixed mapping-based scheme.
Highlights
Bigdata analysis is a vital ingredient in modern internet services such as e-commerce, semantic search, and customer recommendation [1,2,3]
This paper introduces a novel adaptive mapping scheme that provides SLA guarantee for urgent clients by preventing them from being interfered by normal clients
This paper raises several issues such as client-typebased separation and data-type-based separation when heterogeneous storage devices are employed in distributed storage systems
Summary
Bigdata analysis is a vital ingredient in modern internet services such as e-commerce, semantic search, and customer recommendation [1,2,3]. Backup or batch processing clients can access data in a best-effort way without interrupting urgent clients [23,24] Such different requirements are contracted as SLA (Service-Level Agreement) between clients and storage providers, which can be expressed in various forms such as performance, reliability, and cost [25,26,27,28]. This paper proposes a novel resource management scheme for supporting SLA in a distributed storage system It classifies clients into two types: one is urgent clients who require guaranteed performance, and the other is normal clients who can be served in a best-effort way. This paper examines how to exploit heterogeneous storage devices, HDDs (Hard Disk Drives) and SSDs (Solid State Drives), to support SLA It reveals that a heterogeneity-oblivious mapping does not reap the performance benefit of SSDs due to the replication mechanism in a distributed storage system.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
