Replica Placement for Availability in the Worst Case

Gao, Debin; Reiter, Michael K.

doi:10.1109/icdcs.2015.67

Cited by 5 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RF of 1 is usually not advised since, in case of failure, the system may not be able to recover the data. Probabilistic methods have been devised to analyze the influence on RF on the system availability and resilience to malicious users and identify an appropriate assignment [30], [31], [32].…”

Section: A Cluster Sizementioning

confidence: 99%

PAX: Partition-aware autoscaling for the Cassandra NoSQL database

Dipietro

Buyya

Casale

2018

NOMS 2018 - 2018 IEEE/IFIP Network Operations and Management Symposium

View full text Add to dashboard Cite

Apache Cassandra has emerged as one of the most widely adopted NoSQL databases. However, there is still a limited understanding on how to optimally operate Cassandra in the cloud using autoscaling methods, by which resources can be scaled up or down to reduce operational costs and meet servicelevel objectives (SLOs). To address this limitation, we present PAX, a partition-aware elastic resource management system for Apache Cassandra. PAX uses low-overhead query sampling and knowledge of the datapartitioning across the nodes to automatically adapt capacity in Cassandra clusters. Differently from existing autoscaling methods for Cassandra, which incur large acquisition times for new nodes, PAX exploits Cassandra's hinted handoff mechanism and a shared hints storage to minimize the time needed to acquire a node into the cluster. We propose a reactive and a proactive implementation of PAX and compare their performance against different workloads with varying intensities and item popularity distributions, finding that the proactive version significantly reduces SLO violations.

show abstract

Section: A Cluster Sizementioning

confidence: 99%

PAX: Partition-aware autoscaling for the Cassandra NoSQL database

Dipietro

Buyya

Casale

2018

NOMS 2018 - 2018 IEEE/IFIP Network Operations and Management Symposium

View full text Add to dashboard Cite

show abstract

“…A pure point-to-point network has no concept of a client or server, it only has equal peer nodes, and ACTS as a client and server for other nodes on the network. P2P networks can be used for many purposes, such as various file sharing software, real-time media business, etc [5] New features based on cloud computing environment, such as s mass data, super large-scale user scale and high availability, traditional data storage and management pose new challenges. The traditional data storage and management is facing new challenges.The distributed storage system and its related technologies are flexible and adaptable to the complex data storage and management tasks.…”

Section: P2p Storage Technologymentioning

confidence: 99%

Research on Data Storage Technology in Cloud Computing Environment

2018

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Abstract.With the rapid development of information technology, cloud computing, large data, mobile interconnection and other technologies, the number of users is increasing, and the amount of data generated by users is increasing exponentially. How to store large scale and massive data, the traditional distributed storage technology will face challenges. Distributed storage technology can use multiple servers to store data and share storage load. However, in the cloud computing environment, it looks pale. In order to improve the storage efficiency of massive data, data blocking algorithm is studied to shorten the response time of data backup. Effective application of data duplication technology improves the storage performance of data. IntroductionWith the rapid development of cloud computing, big data and mobile Internet technology, cloud computing technology as the core of the new generation of IT technology [1], it is exerting a subtle influence on people's way of life and learning. The concept of cloud computing was first put forward at the search engine conference held in 2006. So far, it has penetrated into all walks of life. Cloud computing is to automatically split a huge computing process into numerous smaller subroutines through a network, and then make a large system composed of multiple servers to assemble a large number of heterogeneous storage devices in the network and return the results to users after search, calculation and analysis. In this environment, the volume of data generated by enterprises and individual users is increasing exponentially. It is estimated that the total data of 40ZB will be generated worldwide by 2020 [2]. How to store large scale, massive data, data processing and storage technology is facing new challenges. Traditional data storage technologies such as network storage, centralized storage, and distributed file systems can not efficiently complete the task of storing and processing mass data in the cloud computing environment.Therefore, this paper makes detailed research on data storage technology under the cloud computing environment, improves the data partitioning strategy, shortens the response time of data backup, and effectively applicates duplicate data deletion technique.to improve the efficiency of data storage. As the core component of the new generation of IT technology, cloud computing will become a new hot spot of the new revolution of information technology in the world after the personal computer and the Internet [3].

show abstract