Swift: a storage architecture for large objects

Cabrera, Luis-Felipe; Long, Darrell D. E.

doi:10.1109/mass.1991.160223

Cited by 8 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typical examples of object storage are AWS S3 (Simple Storage Service) [29], Microsoft Azure [30], and Google Cloud Storage [31], that are provided by cloud service venders, while examples of open source developments include Swift [32], [33], GlusterFS [34], [35], and Ceph [36], [37].…”

Section: Object Storagementioning

confidence: 99%

Survey of Cloud-Based Content Sharing Research: Taxonomy of System Models and Case Examples

Sugawara

2017

IEICE Trans. Commun.

View full text Add to dashboard Cite

SUMMARY This paper illustrates various content sharing systems that take advantage of cloud's storage and computational resources as well as their supporting conventional technologies. First, basic technology concepts supporting cloud-based systems from a client-server to cloud computing as well as their relationships and functional linkages are shown. Second, the taxonomy of cloud-based system models from the aspect of multiple clouds' interoperability is explained. Interoperability can be categorized into provider-centric and client-centric scenarios. Each can be further divided into federated clouds, hybrid clouds, multi-clouds and aggregated service by broker. Third, practical cloud-based systems related to contents sharing are reported and their characteristics are discussed. Finally, future direction of cloud-based content sharing is suggested. key words : content sharing, client-server, shared storage, peer-to-peer, grid computing, cloud computing, object storage, interoperability, federated cloud, hybrid cloud, multi-cloud, aggregated service by broker, provider centric, client centric IntroductionDue to rapid advances of computers, communication devices, and communication networks, diverse and vast amounts of digital contents are exchanged over the Internet, enabling numerous users to enjoy the benefits of technology. To realize smooth and stress-free content provisioning, searching and retrieving, numerous content sharing systems and supporting technologies have been developed and implemented. Currently, cloud computing systems for content sharing, such as exchanging text messages, documents, photos, voice, music, and videos, are important in our daily life, and many interesting systems have been implemented over the Internet.According to an Internet traffic forecast issued by Cisco [1], global IP traffic in 2015 stands at 72.5 exabytes (EB) per month and will nearly triple by 2020, to reach 194.4 EB per month. On the other hand, the percentage of Content Delivery Network (CDN) traffic in global IP traffic is expected to grow rapidly. CDN traffic in 2015 stands at 23.9 EB per month which is about one third of global IP, will grow to 104.0 EB which is more than half of global IP. Cloud-based content sharing plays a major role of CDN, and its importance must be getting larger and larger in the next decade. However, the concept of cloud computing is rather vague even now, and its definition is partially controversial. Then, it is sometimes difficult even for the researchers in the computer and communications fields to truly understand the natures of numerous cloud-based content sharing systems working over the Internet. Thus we would like this survey to be some help for the researchers to organize their ideas or concepts on cloud-based content sharing systems.The rest of this paper is organized into four sections. In Sect. 2, some basic technology concepts supporting cloudbased systems from client-server to cloud computing, including their relationships and functional linkages, are shown. Section 3 int...

show abstract

Section: Object Storagementioning

confidence: 99%

Survey of Cloud-Based Content Sharing Research: Taxonomy of System Models and Case Examples

Sugawara

2017

IEICE Trans. Commun.

View full text Add to dashboard Cite

show abstract

“…In the initially configuration, data models are serialized and stored in cloud-based object storage --OpenStack Swift (Cabrera & Long, 1991) (OpenStack, 2013). It is after found that the RESTful style of accessing (searching, writing, and retrieving) data models are very inefficient, especially when there are a large number of data models stored in a single container (Figure 11).…”

Section: Cloud Projection Transmission Costmentioning

confidence: 99%

Efficient Private Cloud Operation using Proactive Management Service

Dong¹,

Herbert²

2013

STCC

View full text Add to dashboard Cite

Operation management for a private cloud infrastructure faces many challenges including efficient resource allocation, load-balancing, and quick response to real-time workload changes. Traditional manual IT operation management is inadequate for this highly dynamic and complex environment. This work presents a distributed service architecture which is designed to provide an automated, shared, and off-site operation management service for private clouds. The service architecture incorporates important concepts such as: Metric Templates for minimising the network overhead for transmission of cloud metrics; a Cloud Projection that provides a global view of the current status and structure of the cloud, supporting optimal decision making; and a Calendar-based Data Storage Model to reduce the storage required for cloud metric data and increase analysis performance. A proactive response to cloud events is generated based on statistical analysis of historical metrics and predicted usage. The architecture, functional components, and operation management strategies are described. A prototype implementation of the proposed architecture was deployed as a service on the IBM SmartCloud. The effectiveness and usability of the proposed proactive operation management solution has been comprehensively evaluated using a simulated private cloud with dynamic and real-world workloads.

show abstract

“…Based on the actual rates different classes are receiving, the gateway will throttled request streams to attempt to steer the actual sharing to match the desired sharing. Swift [2] is a distributed storage system that also stripes objects across different storage nodes. A QoS mechanism for Swift was proposed [10].…”

Section: Related Workmentioning

confidence: 99%