Multiple Storage Servers Research Articles

Software Defined Network Function Virtualization Framework for Securing Cloud with Data Fusion and Machine Learning Techniques

Computing in the cloud is one of the platforms that may be used to provide distributed computing resources. Supplying and managing cloud resources most effectively is referred to as resource management. A recent development in technology known as fog computing is an example of an expanded and dispersed infrastructure. This architecture maintains application processes between end devices and the network edge to provide more dependable and efficient services. These services include remote data storage, allowing customers to access their data from a distant location. Providing remote storage service is an advantageous function offered by cloud suppliers. On the other hand, the data stored in the cloud is geographically dispersed and kept in various data centers, significantly increasing the risk to users' privacy and security. One of the problems that might arise with privacy is when many data centers store the same information. Many cloud service providers check their customers' data using a Third-Party Auditor (TPA) to address concerns about client privacy and data integrity. Currently, most trusted TPAs only have one validator, making it impossible to expand the data integrity across several data centers. The various verifiers used by TPAs have been reduced in number in response to Man in the Cloud (MiTC) attacks. As a result, they cannot check and authenticate the integrity of data stored in several data centers. A unique Peer to Peer (P2P) authentication protocol with Certificate Authority (CA) and Data Storage Protocol is presented as a solution to the problem that has been outlined above to check for and go around any issues that may arise (DSP). The efficiency of the proposed protocol is demonstrated by the incorporation of TPAs and Certificate Authorities. The proposed protocol has been tested with a single user and a single storage server, as well as multiple storage servers in ownCloud with one backup server, two storage servers, three clients, and two TPAs. The NoSQL server in an organization's cloud is set up to save data to storage servers in the appropriate format. The Amanda backup server is used to back up the mirror copy of the stored data on the storage servers. Automated Validation of Internet Security Protocols with Data Fusion and Applications, or AVISPA for short, is a technology that may be used to verify data stored in the cloud. The findings make it abundantly evident that the suggested protocol is strong enough to guarantee the authenticity of data kept in several data centers.

Enforcing End-to-End I/O Policies for Scientific Workflows Using Software-Defined Storage Resource Enclaves

Data-intensive knowledge discovery requires scientific applications to run concurrently with analytics and visualization codes executing in situ for timely output inspection and knowledge extraction. Consequently, I/O pipelines of scientific workflows can be long and complex because they comprise many stages of analytics across different layers of the I/O stack of high-performance computing systems. Performance limitations at any I/O layer or stage can cause an I/O bottleneck resulting in greater than expected end-to-end I/O latency. In this paper, we present the design and implementation of a novel data management infrastructure called Software-Defined Storage Resource Enclaves (SIREN) at system level to enforce end-to-end policies that dictate an I/O pipeline's performance. SIREN provides an I/O performance interface for users to specify the desired storage resources in the context of in-situ analytics. If suboptimal performance of analytics is caused by an I/O bottleneck when data are transferred between simulations and analytics, schedulers in different layers of the I/O stack automatically provide the guaranteed lower bounds on I/O throughput. Our experimental results demonstrate that SIREN provides performance isolation among scientific workflows sharing multiple storage servers across two I/O layers (burst buffer and parallel file systems) while maintaining high system scalability and resource utilization.

Data placement approach for scalable online social networks

Online social networks are attracting more and more users. Faced with hundreds of millions of users, how to store user data in a scalable manner has become a hot issue of focus for both social service providers and researchers. Currently, distributed key value store is widely used; it places user data across multiple storage servers based on a hash approach. However, it results in a huge amount of communication traffic inside a data center, and is not conducive to the scale of social networks. By considering user interaction characteristics, this paper proposes a data placement approach that combines both social graph partitioning and data replication. Considering the network topologies of data centers, we design the data placement for specific topologies. Furthermore, we discuss the incremental adjustment for social network growth and the distributed implementation of the proposed algorithms. Finally, experiments on real world traces indicate that the proposed algorithms can effectively reduce internal communication traffic, thereby enhancing the scalability of online social networks.

Beehive: Erasure Codes for Fixing Multiple Failures in Distributed Storage Systems

In distributed storage systems, erasure codes have been increasingly deployed to tolerate server failures without loss of data. Traditional erasure codes, such as Reed-Solomon codes, suffer from a high volume of network transfer and disk I/O to recover unavailable data at failed storage servers. Typically, unavailable data at different failed storage servers in a distributed storage system are fixed separately. It has been shown that it is possible to reduce the volume of network transfer significantly by reconstructing data from multiple storage servers at the same time. However, there has been no construction of erasure codes to achieve it without imposing strict constraints on system parameters. In this paper, we propose Beehive codes, designed for optimizing the volume of network transfers to fix the data on multiple failed storage servers. Beehive codes can be constructed over a wide range of system parameters at code rate no more than 0.5, while incurring slightly more storage overhead than Reed-Solomon codes. To achieve the optimal storage overhead as Reed-Solomon codes, we further extend vanilla Beehive codes to MDS Beehive codes, which incurs near-optimal volumes of network transfers during reconstruction. We implement both Beehive and MDS Beehive Codes in C++ and evaluate their performance on Amazon EC2. Our evaluation results have clearly shown that the volume of both network transfers and disk I/O can be conserved by a substantial margin.

Unbreakable distributed storage with quantum key distribution network and password-authenticated secret sharing.

Distributed storage plays an essential role in realizing robust and secure data storage in a network over long periods of time. A distributed storage system consists of a data owner machine, multiple storage servers and channels to link them. In such a system, secret sharing scheme is widely adopted, in which secret data are split into multiple pieces and stored in each server. To reconstruct them, the data owner should gather plural pieces. Shamir’s (k, n)-threshold scheme, in which the data are split into n pieces (shares) for storage and at least k pieces of them must be gathered for reconstruction, furnishes information theoretic security, that is, even if attackers could collect shares of less than the threshold k, they cannot get any information about the data, even with unlimited computing power. Behind this scenario, however, assumed is that data transmission and authentication must be perfectly secure, which is not trivial in practice. Here we propose a totally information theoretically secure distributed storage system based on a user-friendly single-password-authenticated secret sharing scheme and secure transmission using quantum key distribution, and demonstrate it in the Tokyo metropolitan area (≤90 km).

Towards Big Linked Data

In light of the challenges of effectively managing Big Data, the authors are witnessing a gradual shift towards the increasingly popular Linked Open Data (LOD) paradigm. LOD aims to impose a machine-readable semantic layer over structured as well as unstructured data and hence automate some data analysis tasks that are not designed for computers. The convergence of Big Data and LOD is, however, not straightforward: the semantic layer of LOD and the Big Data large scale storage do not get along easily. Meanwhile, the sheer data size envisioned by Big Data denies certain computationally expensive semantic technologies, rendering the latter much less efficient than their performance on relatively small data sets. In this paper, the authors propose a mechanism allowing LOD to take advantage of existing large-scale data stores while sustaining its “semantic” nature. The authors demonstrate how RDF-based semantic models can be distributed across multiple storage servers and the authors examine how a fundamental semantic operation can be tuned to meet the requirements on distributed and parallel data processing. The authors' future work will focus on stress test of the platform in the magnitude of tens of billions of triples, as well as comparative studies in usability and performance against similar offerings.