Search

Distributed storage systems play a pivotal role in modern data-intensive applications, with Hadoop distributed file system (HDFS) being a prominent example. However, optimizing the efficiency of such systems remains a complex challenge. This research paper presents a novel approach to enhance the efficiency of distributed storage by leveraging multi-agent systems (MAS). Our research is centered on enhancing the efficiency of the HDFS by incorporating intelligent agents that can dynamically assign storage tasks to nodes based on their performance characteristics. Utilizing a decentralized decision-making framework, the suggested approach based on MAS considers the real-time performance of nodes and allocates storage tasks adaptively. This strategy aims to alleviate performance bottlenecks and minimize data transfer latency. Through extensive experimental evaluation, we demonstrate the effectiveness of our approach in improving HDFS performance in terms of data storage, retrieval, and overall system efficiency. The results reveal significant reductions in job execution times and enhanced resource utilization, there by offering a promising avenue for enhancing the efficiency of distributed storage systems. [ABSTRACT FROM AUTHOR]

In distributed storage systems, the utilization of locally repairable codes offers the potential to reduce the complexity and bandwidth required for repairs. This paper focuses on a scenario where each information symbol is associated with several distinct repair sets, each of which includes a single parity check symbol. By leveraging various combinatorial designs, such as resolvable balanced incomplete block designs and resolvable group divisible designs, we establish regular packings and resolvable packings for all alphabet sizes greater than a specified constant. These packings enable the construction of optimal locally repairable codes with multiple repair sets. Specifically, in the case where r = 4 , we successfully construct optimal locally repairable codes with multiple repair sets, provided that the code dimension k is greater than or equal to 277. [ABSTRACT FROM AUTHOR]

A new family of scalable codes with locality and availability for information repair in data storage systems for e-health applications was presented recently. The construction was based on a graph of the [7, 3-4] Simplex code. In this paper it is shown that the construction can be generalized via tessellation in a Euclidian plane. The codes obtained have new interesting recoverability properties. They can in some cases repair damage to many storage nodes in multiple connected graphs via sequential decoding, which is similar to healing wounds in biological systems. The advantages of the original codes, namely the availability, functionality, efficiency and high data accessibility, will be preserved also in these new codes. The computational complexity and communication costs of their incrementation will remain constant and modest. These codes could be adapted to disaster recovery because it is straightforward to place the nodes so that the graph is easily mapped on a real structure in space. [ABSTRACT FROM AUTHOR]

The Singleton-type bound that offers an upper limit on the minimum Hamming distance for locally repairable codes (LRCs) has been studied by a lot of literatures. However, the work to determine tight Singleton-type bound for LRCs under all parameters is still incomplete. This paper focuses on the tightness of Singleton-type bound for general LRCs (linear and nonlinear). By exploring how the structure of repair sets affects the minimum Hamming distance, some properties for minimum Hamming distance of LRCs are derived. Through these properties, the following results are obtained: 1. The necessary conditions for LRCs whose minimum Hamming distance reaches the previous Singleton-type bound are provided. Due to these necessary conditions, the previous Singleton-type bound is proved to be unreachable under a number of parameters, and the tight Singleton-type bounds are given under some of these parameters. 2. Under a class of parameters where the previous Singleton-type bound for linear LRCs has been proved unreachable in the literatures, we improve Singleton-type bound and show that the new Singleton-type bound is tight. In addition, based on the existing LRC constructions, the previous Singleton-type bound for LRCs is shown to be tight under some other parameters. Finally, for an importance case of LRCs (i.e., every repair set is a 1-erasure correcting code), the Singleton-type bound under three classes of parameters is further improved. [ABSTRACT FROM AUTHOR]

In the field of distributed medical data storage, relayed data is usually kept centrally on the central node, which may lead to an increased risk of a single point of failure and security challenges from malicious tampering. While introducing backup nodes can partially alleviate these issues, the efficiency of synchronization and switching between nodes remains constrained. In addition, nodes storing relayed data can reach a consensus to modify the forwarded data, which lacks trustworthiness. Given the problems faced by traditional distributed storage in the field of medical data, combined with the characteristics of blockchain technology, we propose a distributed medical data storage model based on blockchain technology, named DDS (Decentralized Distributed Storage), which is intended to be applied in medical data management. In the DDS model, we embed relay data into blocks for storage and ensure the integrity and security of relay data for medical data by introducing redundant blockchain storage and collaborative verification mechanisms. The model is divided into two key phases: the storage phase of relay data of medical data and the verification phase. In the storage phase, the doctor's signature and the associated copy location data are sent to multiple validation nodes, after which the relay data block of medical data is generated and written to the blockchain. In the validation phase, the node first verifies whether the local relay data blockchain state is consistent with the global state. If it is not, a state synchronization operation is performed. Next, the validation node retrieves the local relay data blockchain to verify the integrity of the medical data relay data. From the theoretical analysis and experimental results, the DDS model can ensure the traceability and integrity of relay medical data and has good concurrent processing performance and less impact on the efficiency of medical data storage. [ABSTRACT FROM AUTHOR]

The increased digitalization of society raises concerns regarding data protection and user privacy, and criticism on how the companies handle user data without being transparent and without providing adequate mechanisms for users to control how their own data is being processed or shared. To address this problem and open the way for a secure and efficient society, where the privacy of citizens is paramount, the identity concept and proof of identity mechanisms need to be redesigned from the ground up. In this paper we discuss how the emerging Web3 technologies like distributed ledger technology (DLT), blockchain, smart contracts, decentralized storage systems, and crypto wallets can be leveraged to design and implement a decentralized digital identity system based on decentralized identifiers (DID) and self-sovereign identities (SSI). Such a system puts the users in full control over their own data while also providing a solid backbone for building interoperable systems that are secure, scalable, and efficient. We propose different architectures for the decentralized identity infrastructure and storage layer, and also discuss the mapping of these architectures on cloud platforms. The main goal is to provide an architectural blueprint for a scalable, secure, privacy-preserving and trusted system.

As the modern computing market experiences a surge in demand for efficient data-management solutions, challenges posed by centralized storage systems become more pronounced, especially with the proliferation of Internet of Things devices. Centralized storage, although cost-effective, faces issues of scalability, performance bottlenecks, and security vulnerabilities. With decentralized storage, data are distributed across nodes, offering redundancy, data availability, and enhanced security. Unfortunately, decentralized storage introduces its own challenges, such as complex data retrieval processes, potential inconsistencies in data versions, and difficulties in ensuring data privacy and integrity in a distributed setup. Effectively managing these challenges calls for innovative techniques. In response, this paper introduces a decentralized storage system that melds cloud-native concepts with blockchain technology. The proposed design delivers enhanced scalability, data security, and privacy. When operating on a containerized edge infrastructure, this storage system provides higher data-transfer speeds than the interplanetary file system. This research thus blends the advantages of cloud-native frameworks with the security mechanisms of blockchain, crafting a storage system that addresses the present-day challenges of data management in decentralized settings.