Bristrita: Namespace and Metadata Distribution in Large-Scale Distributed Cloud Storage Systems

Author(s):  
Hrishikesh Dewan ◽  
Ramesh Hansdah
Author(s):  
Sebastian Dippl ◽  
Michael C. Jaeger ◽  
Achim Luhn ◽  
Alexandra Shulman-Peleg ◽  
Gil Vernik

While it is common to use storage in a cloud-based manner, the question of true interoperability is rarely fully addressed. This question becomes even more relevant since the steadily growing amount of data that needs to be stored will supersede the capacity of a single system in terms of resources, availability, and network throughput quite soon. The logical conclusion is that a network of systems needs to be created that is able to cope with the requirements of big data applications and data deluge scenarios. This chapter shows how federation and interoperability will fit into a cloud storage scenario. The authors take a look at the challenges that federation imposes on autonomous, heterogeneous, and distributed cloud systems, and present approaches that help deal with the special requirements introduced by the VISION Cloud use cases from healthcare, media, telecommunications, and enterprise domains. Finally, the authors give an overview on how VISION Cloud addresses these requirements in its research scenarios and architecture.


Author(s):  
Amina Mseddi ◽  
Mohammad Ali Salahuddin ◽  
Mohamed Faten Zhani ◽  
Halima Elbiaze ◽  
Roch H. Glitho

2019 ◽  
Vol 2 (2) ◽  
pp. 158-169
Author(s):  
Ha Huy Cuong Nguyen ◽  
◽  
Tung Trong Nguyen ◽  
Trung Hai Trinh

2017 ◽  
Vol 14 (11) ◽  
pp. 82-91 ◽  
Author(s):  
Jing Wang ◽  
Wei Luo ◽  
Wei Liang ◽  
Xiangyang Liu ◽  
Xiaodai Dong

2012 ◽  
Vol 433-440 ◽  
pp. 5861-5865
Author(s):  
Zhen Huang ◽  
Yuan Yuan ◽  
Yu Xing Peng

The ever-growing demand on information and data requires the efficient architecture for large-scale network storage systems. To serve very large scale applications, using inexpensive commodity becomes the common selection in nowadays cloud storage systems. Based on such unreliable hardware, building fault-tolerant mechanism is key issue to the system design. In this paper, we propose a rack-aware architecture for cloud storage systems.


IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 108148-108157
Author(s):  
Hailiang Xiong ◽  
Changwu Hu ◽  
Yujun Li ◽  
Guangyuan Wang ◽  
Hongchao Zhou

Author(s):  
Anthony Kougkas ◽  
Hassan Eslami ◽  
Xian-He Sun ◽  
Rajeev Thakur ◽  
William Gropp

Key–value stores are being widely used as the storage system for large-scale internet services and cloud storage systems. However, they are rarely used in HPC systems, where parallel file systems are the dominant storage solution. In this study, we examine the architecture differences and performance characteristics of parallel file systems and key–value stores. We propose using key–value stores to optimize overall Input/Output (I/O) performance, especially for workloads that parallel file systems cannot handle well, such as the cases with intense data synchronization or heavy metadata operations. We conducted experiments with several synthetic benchmarks, an I/O benchmark, and a real application. We modeled the performance of these two systems using collected data from our experiments, and we provide a predictive method to identify which system offers better I/O performance given a specific workload. The results show that we can optimize the I/O performance in HPC systems by utilizing key–value stores.


Sign in / Sign up

Export Citation Format

Share Document