Latency-Sensitive Data Allocation and Workload Consolidation for Cloud Storage

Yang, Song; Wieder, Philipp; Aziz, Muzzamil; Yahyapour, Ramin; Fu, Xiaoming; Chen, Xu

doi:10.1109/access.2018.2883674

Cited by 8 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the storage is matched to check whether it meets the requirement of the recent file. en, the suitable location is selected to limit or minimize the latency issue [16].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Fast Data Access through Nearest Location-Based Replica Placement

Shoaib

Arshad

Khattak

et al. 2022

Scientific Programming

View full text Add to dashboard Cite

Nowadays, millions of telecommunication devices, IoT sensors, and web services, especially social media sites, are producing big data every second. Such applications with massive data-generating capabilities need to access these data quickly. Among other approaches, cloud computing provides content delivery networks, which utilize data replication for better latency to such real-time applications. Fast processing, storage, and timely analysis of these data are the challenge for most of these future Internet applications. However, cloud computing is the ultimate storage and processing paradigm to resolve such issues and to deal with big data, with the same speed in which data are being produced. Moreover, cloud computing technology has been evolved with the tag line of “Everything as A Service” now, and for powering all these services, provision of data is a compulsory and significant task. To provide end user with easy and fast access to data, the cloud maintains backup and replicates the copies in multiple data centers. The geographical locations of these data centers where data are being placed have a profound impact on data access time. To deal with the challenge of effective data replication and to reduce data access time to a minimum, we propose a genetic algorithm (GA)-based technique to suggest and store data on nearly located data centers. The proposed algorithm improves the access time, and thus, the efficiency of cloud servers by providing quality of service (QoS) to end user.

show abstract

“…In addition, the storage is matched to check whether it meets the requirement of the recent file. en, the suitable location is selected to limit or minimize the latency issue [16].…”

Section: Related Workmentioning

confidence: 99%

“…Increasing performance of data access in distributed systems is using replication [15]. Replicating multiple copies of files in different places increases the performance of data access and minimizing response time [16].…”

Section: Introductionmentioning

confidence: 99%

Fast Data Access through Nearest Location-Based Replica Placement

Shoaib

Arshad

Khattak

et al. 2022

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…As the continuous expansion of the network scale, the rapid increase of Internet devices, and the explosive growth of data, only rely on cloud storage and cloud computing cannot meet the client's demand for high-performance and real-time services. Therefore, the edge-cloud collaborative architecture has become a feasible solution [1]. The edge of the edge-cloud collaborative architecture adopts a distributed multi-copy storage method to ensure high fault tolerance and high availability of the system.…”

Section: Introductionmentioning

confidence: 99%

CCECGP: causal consistency model of edge–cloud collaborative based on grouping protocol

2022

View full text Add to dashboard Cite

At present, most causal consistency models based on cloud storage can no longer meet the needs of delay-sensitive applications. Moreover, the overhead of data synchronization between replicas is too high. This paper proposes a causal consistency model of edge-cloud collaborative based on grouping protocol. The model based on the edge-cloud collaboration architecture, partitions cloud data centers and groups edge nodes by distributed hash tables, and stores a subset of the complete data set in nodes located at the edge of the network. Thereby realize partial geo-replication in edge-cloud collaboration environment. At the same time, we design a group synchronization algorithm called Imp_Paxos, so that the update only needs to be synchronized to the main group, which reduces the visibility delay of the update and decreases the data synchronization overhead.Besides, a sort timestamp is proposed in this paper, which generates different timestamps according to the type of update to track causality, keeping the amount of metadata managed in a relatively stable and low state.Threrfore, the proposed model reduces the overhead of metadata for system management, and improves throughput quantity of system. Experiments show that, our model performs well in terms of throughput, operation latency, and update visibility latency compared with existing causal consistency models.

show abstract

“…In addition, with the extensive use of deep learning [5] and augmented reality (AR) technologies [6], i.e., automatic driving [7] and smart home [8], terminal devices needs the ability of realtime data processing. However, traditional end-cloud architecture adopts centralized management, in which all data flows to the central-cloud (CC), which causes great overhead of network bandwidth and makes congestion in CC [9]. Moreover, CC may require data transmitted far across geographical distances with great delay, which is difficult to meet the realtime requirements.…”

Section: Introductionmentioning

confidence: 99%

Grouping-Based Consistency Protocol Design for End-Edge-Cloud Hierarchical Storage System

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

With the increasing of the number of edge-devices and the demand of real-time experience, the end-edge-cloud hierarchical storage system (EECHSS) is emerged recently for reliable caching and fast offloading of massive amounts of data. EECHSS can accommodate various services, save computing power and improve storage capacity, due to transformation from central-cloud to edge-cloud, considerable reducing service delay and communication overhead. One of the main challenges brought by edge-cloud architecture is consistency problem. It is difficult to guarantee that the data from the two distributed clusters is consistent by existing consistency protocols. In addition, as the variety of applications increases, the existing fixed consistency level of classical protocols can no longer satisfy the system dynamic requirements. In this paper, we focus on designing grouping-based consistency protocols with adaptively selecting consistency level in EECHSS. At first, we analyze the internal structure and the workflow of EECHSS, and devise two modified adaptive grouping-based consistency protocols (GM-Paxos and GEPaxos) with efficient grouping algorithms. Then, for the characteristics that data is offloaded frequently, we design two synchronization strategies to ensure the consistency of the data cached in the edge-cloud and the central-cloud, respectively. Experiments show that, our proposed grouping-based consistency protocols of EECHSS can improve the availability as much as possible while ensuring data consistent. INDEX TERMS End-edge-cloud hierarchical storage system, adaptive consistency, grouping algorithm, synchronization strategy.

show abstract

Latency-Sensitive Data Allocation and Workload Consolidation for Cloud Storage

Cited by 8 publications

References 28 publications

Fast Data Access through Nearest Location-Based Replica Placement

Fast Data Access through Nearest Location-Based Replica Placement

CCECGP: causal consistency model of edge–cloud collaborative based on grouping protocol

Grouping-Based Consistency Protocol Design for End-Edge-Cloud Hierarchical Storage System

Contact Info

Product

Resources

About