Segank: A Distributed Mobile Storage System

Concurrency and Computation: Practice and Experience, 2007

The rising demand for mobile computing has created a need for an improved file system that supports mobile clients. Current file systems with support for mobility provide availability through file replicas that are cached at the client side. However, mobile clients may experience different obstacles with regards to the local cache, such as the limited network bandwidth, the intermittent connection, and serious conflicts when synchronizing back to the server. In this paper, we propose a novel mobile distributed file system design, which provides high availability and reliable storage for files and guarantees that file operations are executed regardless of concurrency and failure issues. The design is intended to fit mobile clients (e.g. PDAs and cell phones) that have limited storage space and cannot store all of the data they need, and yet require access to these data at all times. We adopt a server-side caching in order to guarantee sufficient caching space to all mobile clients, and to ensure the availability of files in the case of clients' failures. We present our algorithm, describe its implementation, simulate its high availability functions, and report on its performance evaluation using a cluster of workstations. Our simulation results indicate clearly that our algorithm exhibits a significant degree of automation and conflict-free mobile file system.

2002

This paper presents the design and implementation of a mobile storage system called a Personal-RAID. PersonalRAID manages a number of disconnected storage devices. At the heart of a Personal-RAID system is a mobile storage device that transparently propagates data to ensure eventual consistency. Using this mobile device, a PersonalRAID provides the abstraction of a single coherent storage name space that is available everywhere, and it ensures reliability by maintaining data redundancy on a number of storage devices. One central aspect of the PersonalRAID design is that the entire storage system consists solely of a collection of storage logs; the log-structured design not only provides an efficient means for update propagation, but also allows efficient direct I/O accesses to the logs without incurring unnecessary log replay delays. The PersonalRAID prototype demonstrates that the system provides the desired transparency and reliability functionalities without imposing any serious performance penalty on a mobile storage user.

Peer-to-peer storage systems organize a collection of symmetric nodes to store data without the use of centralized data structures or operations. As a result, they can scale to many thousands of nodes and can be formed directly by clients. This paper describes the design and implementation Mammoth, which implements a traditional UNIX-like hierarchical file system in a peer-to-peer fashion. Each Mammoth node stores a potentially arbitrary collection of directories and files that may be replicated on multiple nodes. The only thing that links these nodes together is that each metadata object encodes the network addresses of the nodes that store it. Data is replicated by a background process whose operation is simplified by the fact that files are stored as journals of immutable versions. An optimistic replication technique is used to allow nodes to read and write whatever version of data they can access, while also ensuring consistency when nodes are connected. In the event of temporary failure, eventual consistency is achieved by ensuring that every replica of a directory or file metadata object receives all updates to the object, irrespective of delivery order. While an update is being propagated every node that receives it cooperates to ensure that the update is delivered, even if the original sender fails. Our prototype is implemented as a user-level NFS server. Its performance is comparable to a standard NFS server and it will be publicly available soon.

19th IEEE International Parallel and Distributed Processing Symposium

Mobile Peer-to-Peer (MP2P) systems consist of mobile peers that collaborate with each other to complete application problems. Information sharing in such environments is a challenging problem due to the fundamental limitations of battery power, wireless bandwidth, and users' frequent mobility. We proposed a novel scheme, called Proximity Regions for Caching in Cooperative MP2P Networks (PReCinCt) to efficiently support scalable data retrieval in large-scale MP2P networks. In the PReCinCt scheme, the network topology is divided into geographical regions where each region is responsible for a set of keys representing the data. In this paper, we extend the PReCinCt scheme to facilitate consistent cooperative caching in MP2P systems. The caching scheme considers data popularity, data size and region-distance during replacement to optimize cache content of peers. PReCinCt employs a hybrid push/pull mechanism to maintain data consistency among replicas in the network. Simulation results show the cost of consistency maintenance in terms of latency and energy consumption is significantly improved in the PReCinCt scheme.

Operating Systems Review, 2001

This paper presents and evaluates the storage management and caching in PAST, a large-scale peer-to-peer persistent storage utility. PAST is based on a self-organizing, Internetbased overlay network of storage nodes that cooperatively route file queries, store multiple replicas of files, and cache additional copies of popular files.

Due to the booming growth of Information and Communication Technology (ICT), a vast amount of data is produced at a considerably high rate and it drives the traditional methods of storing data to its limits and most of the time it simply overwhelms the current storage systems. Because of that throughout the history of the development of ICT, the effort to find an efficient and feasible data storage system that has a substantial capacity to cater the current data storage needs has been relentless.Currently available shared storage devices are mostly file servers and Peer-to-Peer systems which are organized in various different architectures but there are certain areas that pose problems in implementing such systems at small scale and also at enterprise level. Networked Shared Storage System(NSS) is introduced as a system motivated by that historical desire to achieve the ultimate reliable and secure storage media and it represents the way ahead in discovering the ultimate solution for this long lasting problem. NSS is a Local Area Network (LAN) based, secure and reliable distributed storage system. Its primary objective is to use the free local hard disk space available in the workstations connected to a LAN, as its storage media. This is achieved by the radical but fail proof method of breaking down the single file in to a set of data chunks and distributing them throughout the LAN. These chunks are then remerged to reproduce the original file at the users’ request. Through this method the largely the unutilized free disk space of nodes connected to the LAN is used to create a free disk space pool that will serve the storage needs of the users of that same network, rather than incorporating separate data servers for that. A LAN based storage system is invariably challenged by the inherent unavailability of the nodes of a LAN. But the NSS overcomes this problem via a robust and efficient data replication algorithm that makes replicas of the data chunks when storing them. Thus providing a high percentage of availability/reliability for the stored data.Peer-to-Peer communication is used when distributing the chunked data throughout the network via the embedded FTP servers. This architecture will minimize the security issues and protect the privacy of data which is greatly challenged in a LAN based environment. NSS is highly scalable and applicable for both medium scale LAN and its enterprise level equivalent, with no additional modification to the architecture and with lesser cost and effort than most existing solutions (Cloud Servers and Server Farms). Thus making it the way ahead in achieving the ultimate storage m

2004

Peer-to-peer file sharing applications have become increasingly popular. Measurements of P2P systems indicate large heterogeneity in the availability of individual nodes. Many have cyclic behavior, whereas others are always available. This paper proposes a cooperative storage technique which employs erasure coding schemes on a collection of data objects and provides various levels of data redundancy. Based on this technique, we study a historybased hill climbing scheme that takes advantage of varied time zones in a global p2p system. Our simulation results show the improved data availability by this scheme. We also investigate several climbing strategies including choice of coding schemes and laziness of data movement.

… of the 6th USENIX Conference on File …, 2008

TierStore is a distributed filesystem that simplifies the development and deployment of applications in challenged network environments, such as those in developing regions. For effective support of bandwidth-constrained and intermittent connectivity, it uses the Delay Tolerant Networking store-and-forward network overlay and a publish/subscribe-based multicast replication protocol. TierStore provides a standard filesystem interface and a single-object coherence approach to conflict resolution which, when augmented with application-specific handlers, is both sufficient for many useful applications and simple to reason about for programmers. In this paper, we show how these properties enable easy adaptation and robust deployment of applications even in highly intermittent networks and demonstrate the flexibility and bandwidth savings of our prototype with initial evaluation results.

Advances in Informatics, 2005

Structured peer-to-peer overlay networks or Distributed Hash Tables (DHTs) are distributed systems optimized for storage and retrieval of read-only data. In this paper we elaborate on a method that allows them to manage mutable data. We argue that by altering the retrieval algorithm of DHTs, we can enable them to cope with data updates, without sacrificing their fundamental properties of scalability and fault-tolerance. We describe in detail and analyze an implementation of a Kademlia network capable of handling mutable data. Nevertheless, the corresponding protocol additions can easily be applied to any DHT design. Experimental results show that although the process of managing and propagating data updates throughout the network adds up to the total cost of the lookup operation, the extra network utilization can be exploited in favor of overlay resilience to random node joins and failures.

… " of Bucharest Scientific Bulletin, Series C: …, 2010

Reţelele peer-to-peer sunt sisteme distribuite ideale pentru a stoca date replicate, în scopul de a le putea obţine rapid şi transparent, orideunde am dori. O problemă importantă, când vine vorba de aceste sisteme distribuite, este replicarea datelor critice într-o manieră persistentă. Această problemă o adresăm prin lucrarea de faţă, propunând un mecanism suplimentar de asigurare a redundanţei, în plus faţă de mecanismul natural al DHT-ului Chord, de replicare pe un număr fix de succesori. Deşi am prevăzut pentru cercetarea viitoare rularea unor teste intensive de stres pentru prototipul curent, chiar şi rezultatele obţinute până în prezent par să confirme capabilităţile de a adapta gradul de redundanţă datelor critice atunci când e necesar sau la cerere.