Abstract:Cloud federation enables service providers to collaborate to provide better services to customers. For cloud storage services, optimizing customer object placement for a member of a federation is a real challenge. Storage, migration, and latency costs need to be considered. These costs are contradictory in some cases. In this article, we modeled object placement as a multi-objective optimization problem. The proposed model takes into account parameters related to the local infrastructure, the federated environ… Show more
“…69 When a replication strategy is centralized, a central authority controls all aspects of replication. 39,58,65,[70][71][72] Decentralized replication strategies encourage no central control. These strategies are more effective in responding to system changes and dealing with failures and outages situations.…”
Section: Centralized Versus Decentralizedmentioning
confidence: 99%
“…These issues are important because they influence task execution time, storage space usage, network usage and so forth. 20,27,59,61,63,64,72,83,86 Interconnected cloud strategies are more focused on the replicas placement issue since they decide replica locations among the available resources of multiple cloud providers.…”
Section: Addressed Replication Issuesmentioning
confidence: 99%
“…This results in increasing the profit for the provider. 20,63,64,72,73,75,118 Most of the replication strategies are oriented towards the provider since they consider the provider to be the entity responsible for the replication process. Only a few strategies are consumer-centric.…”
Section: Provider-oriented Versus Consumer-orientedmentioning
confidence: 99%
“…20,30,63,118 Provider-centric strategies in the interconnected cloud environments focus mainly on taking advantage of the differences between the pricing policies of the cloud providers among each other to reduce the cost of management. 40,66,72,73,75,101…”
“…Economic cost models are generally related to the economic aspect of the cloud, including the pricing policies, the monetary cost of management, auctions and so forth. 20,30,40,64,66,72,73,78,90,91,118,122,133 The focus on the economic aspect of cloud systems by replication strategies is mainly done by addressing the monetary cost of replication. 20,30,40,64,66,72,73,78,90,91 This monetary cost is often estimated by economic cost models that take into account both the cost of data storage and data transfer as well as the cost of the SLA penalties.…”
SummaryData replication is a well‐known technique in cloud systems for enhancing availability and performance. Various strategies and surveys have been proposed in this respect. These surveys include comprehensive analysis and classifications. However, to the best of the authors' knowledge, there is no survey concentrating on strategies designed for interconnected cloud systems. In this article, we provide an in‐depth analysis of existing data replication strategies in cloud systems, covering single and interconnected clouds. We also highlight data correlation‐aware strategies as well as their key steps. Furthermore, we examine the major strategies' features such as: () addressed replication issues, () orientation towards the provider and the consumer, () consideration of the service level agreement, () consideration of cost and economic aspects, and () evaluation tools. Finally, we provide a performance analysis through extensive simulations of several replication strategies dedicated for single and interconnected clouds.
“…69 When a replication strategy is centralized, a central authority controls all aspects of replication. 39,58,65,[70][71][72] Decentralized replication strategies encourage no central control. These strategies are more effective in responding to system changes and dealing with failures and outages situations.…”
Section: Centralized Versus Decentralizedmentioning
confidence: 99%
“…These issues are important because they influence task execution time, storage space usage, network usage and so forth. 20,27,59,61,63,64,72,83,86 Interconnected cloud strategies are more focused on the replicas placement issue since they decide replica locations among the available resources of multiple cloud providers.…”
Section: Addressed Replication Issuesmentioning
confidence: 99%
“…This results in increasing the profit for the provider. 20,63,64,72,73,75,118 Most of the replication strategies are oriented towards the provider since they consider the provider to be the entity responsible for the replication process. Only a few strategies are consumer-centric.…”
Section: Provider-oriented Versus Consumer-orientedmentioning
confidence: 99%
“…20,30,63,118 Provider-centric strategies in the interconnected cloud environments focus mainly on taking advantage of the differences between the pricing policies of the cloud providers among each other to reduce the cost of management. 40,66,72,73,75,101…”
“…Economic cost models are generally related to the economic aspect of the cloud, including the pricing policies, the monetary cost of management, auctions and so forth. 20,30,40,64,66,72,73,78,90,91,118,122,133 The focus on the economic aspect of cloud systems by replication strategies is mainly done by addressing the monetary cost of replication. 20,30,40,64,66,72,73,78,90,91 This monetary cost is often estimated by economic cost models that take into account both the cost of data storage and data transfer as well as the cost of the SLA penalties.…”
SummaryData replication is a well‐known technique in cloud systems for enhancing availability and performance. Various strategies and surveys have been proposed in this respect. These surveys include comprehensive analysis and classifications. However, to the best of the authors' knowledge, there is no survey concentrating on strategies designed for interconnected cloud systems. In this article, we provide an in‐depth analysis of existing data replication strategies in cloud systems, covering single and interconnected clouds. We also highlight data correlation‐aware strategies as well as their key steps. Furthermore, we examine the major strategies' features such as: () addressed replication issues, () orientation towards the provider and the consumer, () consideration of the service level agreement, () consideration of cost and economic aspects, and () evaluation tools. Finally, we provide a performance analysis through extensive simulations of several replication strategies dedicated for single and interconnected clouds.
SummaryScaling up large‐scale scientific applications on supercomputing facilities is largely dependent on the ability to scale up efficiently data storage and retrieval. However, there is an ever‐widening gap between I/O and computing performance. To address this gap, an increasingly popular approach consists in introducing new intermediate storage tiers (node‐local storage, burst‐buffers,) between the compute nodes and the traditional global shared parallel file‐system. Unfortunately, without advanced techniques to allocate and size these resources, they remain underutilized. In this article, we investigate how heterogeneous storage resources can be allocated on an high‐performance computing platform, just like compute resources. To this purpose, we introduce StorAlloc, a simulator used as a testbed for assessing storage‐aware job scheduling algorithms and evaluating various storage infrastructures. We illustrate its usefulness by showing through a large series of experiments how this tool can be used to size a burst‐buffer partition on a top‐tier supercomputer by using the job history of a production year.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.