Search

Infrastructure-As-A-Service (IAAS) provides an environmental setup under any type of cloud. In Distributed file system (DFS), nodes are simultaneously serve computing and storage functions; that is parallel Data Processing and storage in cloud. Here, file is considered as a data or load. That file is partitioned into a number of File chunks (FC) allocated in distinct nodes so that Map Reduce tasks can be performed in parallel over the nodes. Files and Nodes can be dynamically created, deleted, and added. This results in load imbalance in a distributed file system; that is, the file chunks are not distributed as uniformly as possible among the Chunk Servers (CS). Emerging distributed file systems in production systems strongly depend on a central node for chunk reallocation or Distributed node to maintain global knowledge of all chunks. This dependence is clearly inadequate in a large-scale, failure-prone environment because the central load balancer is put under considerable workload that is linearly scaled with the system size, it may thus become the performance bottleneck and the single point of failure and memory wastage in distributed nodes. So, we have to enhance the Client side module with server side module to create, delete and update the file chunks in Client Module. And manage the overall private cloud and apply dynamic load balancing algorithm to perform auto scaling options in private cloud. In this project, a fully distributed load rebalancing algorithm is presented to cope with the load imbalance problem.

With the development of information technology, distributed file system is widely used in massive information storage. Usually, distributed file system uses metadata server to achieve quick access to files according to directory, thus the organization and management of metadata are the keys to the file system performance. In general, directory subtree partition method and hash algorithm are used by existing mass storage system to manage metadata. However, to solve the problems, like low access efficiency of metadata, ineffective balance of load and poor extensibility, in existing metadata management strategy of distributed file system, dynamic load balance strategy of metadata based on hash tags is put forward, in which the tags act as partition granularity and hot tags of metadata will be hashed again to achieve the goal of load balance. The experimental results in this paper turn out that modified metadata management strategy based on hash tags has greater system throughput and less average response time than the one based on tags.

File Allocation Table (FAT)file system is the most common file system used in embedded devices such as smart phones, digital cameras, smart TVs, tablets, etc. Typically these embedded devices use Solid State Drives (SSD) as storage devices. The ExFAT file system is future file system for embedded devices and it is optimal for SSDs. This paper discourses the methodologies for Geotagging as a file system metadata instead of file data in FAT and ExFAT file systems. The designed methodologies of this paper adheresthe compatibility with the FAT file system specification and existing ExFAT file system implementations. Keywords Cluster, Contiguous, ExFAT, EXIF, FAT, File system, Flash memories, GPS, MMC, Multimedia, Micro SD, NAND, NOR, SSD,Storage, Video, XMP. 1. INTRODUCTION The Multimedia Cards (MMC) [1]/ Secure Digital (SD) [2] / Micro SD cards are composed of NOR and/or NAND flash memories [3]. The NOR and NAND flash memories are also called as “Solid State Drives”. The Flash memories are default choice of any embedded device as they are low-priced, smaller size and higher storage capacity.The FAT [4] file system is the widely adopted file system in SSDs. The multi-media applications such as video imaging, audio/video playback and recording uses FAT file system for storing and retrieving the data of the multimedia files. The first version of FAT file system was FAT12 by Microsoft Corporation, later it was extended as FAT16 and further as FAT32 to support higher storage capacity. The FAT file system was initially developed to use on floppy disks and Hard Disk Drive (HDD) s. Since most of the operating systems of Personal Computer (PC) s such as Windows, Linux, Mac OS, etc. implement the FAT file system, this file system has become a default and world-wide compatible storage format for embedded devices. Usually the device with the implementation of FAT is recognized as removable storage media in a PC. Even though FAT file system does not define flash management techniques such as Wear-Levelling and Bad Block management [4], the embedded devices implements this file system along with the dedicated flash block management algorithms. In FAT File system, the storage device is represented as group of linear clusters. A cluster is basic unit of data storage in FAT file system and it is a group of blocks or sectors of storage device. In FAT file system, the file or directory is a linked list of the clusters. The File Allocation Table stores the status of all available clusters of the device; Status of a cluster can be either allocated as part of file/directory data or free or reserved. An every entry of File Allocation Table indicates the status of corresponding cluster. The FAT file system specification limits the maximum supported storage size to 32 GB. But, the maximum size supported by FAT32 implementation is 128 GB. But, today the flash storage cards of more than 32 GB are available in market. The ExFAT [5] [6] file system is developed, by Microsoft Corporation, as successor of FAT32 file system. This file system is optimally designed to support large size flash storage cards with higher read and write performance. The maximum storage size supported by ExFAT file system in 128 peta bytes (PB). Global PositioningSystem (GPS) based positioning [7][8]is the positioning technique commonly used inSmartphones. Most of Smartphone contains the GPS receiver which continuously receives the signals from the satellites. These satellite signals contain the exacttime of the message sent, and the location information. The GPS receiver usesthe received signals of four or more satellites to determine the current position based on trilateration.Many smart phones uses the multiple sensors improve the accuracy of the GPS position [9]. In Android based mobile phones, Location Manger [10] can be used for the Geotagging of multimedia files. The eXtensible Metadata Platform (XMP) [11] and Exchangeable image file format (EXIF) [12] are the formats used to store the geographical coordinates and other related information in multimedia and Portable Document Format (PDF) files. In both XMP and EXIF formats, the Geographical data is always placed as multimedia file data. Due to this, file need to be opened and read operation is required to retrieve the geographical location of the file. This paper designs the techniques to place the geographical data of the file as attribute of the file instead of user data of the file. In FAT and ExFAT file systems, the attribute of the file contains the file name, size in bytes, date and time of creation and last write. The attribute of the file is referred as Meta data of the file. This paper includes the geographical data to the attribute of the file/directory by following methodologies. i) Geotagging with Reserved Sectors of FAT and ExFAT file systems. ii) Geotagging with Reserved Clusters of FAT file systems. The files created/updated with above techniques adheres the compatibility with the existing FAT and ExFAT file system implementations and specifications. This means, files created with Geographical data as attributes are accessible for read and write operations in other FAT and ExFAT file system implementations without Geotagging facility. The Geotagging methodologies described in this paper stores the geographical

SUMMARY The numbers of files in file systems have increased dramatically in recent years. Office workers spend much time and effort searching for the documents required for their jobs. To reduce these costs, we propose a new method for recommending files and operations on them. Existing technologies for recommendation, such as collaborative filtering, suffer from two problems. First, they can only work with documents that have been accessed in the past, so that they cannot recommend when only newly generated documents are inputted. Second, they cannot easily handle sequences involving similar or differently ordered elements because of the strict matching used in the access sequences. To solve these problems, such minor variations should be ignored. In our proposed method, we introduce the concepts of abstract files as groups of similar files used for a similar purpose, abstract tasks as groups of similar tasks, and frequent abstract workflows grouped from similar workflows, which are sequences of

Current peer-to-peer (P2P) file sharing methods in mobile ad hoc networks (MANETs) can be classified into three groups: flooding-based, advertisement-based, and social contact-based. The first two groups of methods can easily have high overhead and low scalability. They are mainly developed for connected MANETs, in which end-to-end connectivity among nodes is ensured. The third group of methods adapts to the opportunistic nature of disconnected MANETs but fails to consider the social interests (i.e., contents) of mobile nodes, which can be exploited to improve the file searching efficiency. In this paper, we propose a P2P content-based file sharing system, namely SPOON, for disconnected MANETs. The system uses an interest extraction algorithm to derive a node's interests from its files for content-based file searching. For efficient file searching, SPOON groups common-interest nodes that frequently meet with each other as communities. It takes advantage of node mobility by designating stable nodes, which have the most frequent contact with community members, as community coordinators for intracommunity searching, and highly mobile nodes that visit other communities frequently as community ambassadors for intercommunity searching. An interest-oriented file searching scheme is proposed for high file searching efficiency. Additional strategies for file prefetching, querying-completion, and loop-prevention, and node churn consideration are discussed to further enhance the file searching efficiency. We first tested our system on the GENI Orbit testbed with a real trace and then conducted event-driven experiment with two real traces and NS2 simulation with simulated disconnected and connected MANET scenarios. The test results show that our system significantly lowers transmission cost and improves file searching success rate compared to current methods.

In cloud storage system, when we search similar documentation files, keyword-based similarity evaluation scheme is well performed. However, if we want to find similar binary files then it is very difficult to satisfy user request. Because there is no widely used binary file search system that supports similarity evaluation among files. File similarity evaluation is essential for digital forensic and data deduplication field. In the file similarity processing time, the CPU consumption and resource overhead of memory are increased as the number of files increase. Moreover, as the file size is getting bigger, the overhead of metadata management is critical. In this paper, we suggest the similarity evaluation scheme using a hybrid chunking which reduce overall processing time of similarity evaluation. Experiment result shows that the proposed system can reduce processing time and data storage capacity effectively.

Summary As peer-to-peer networks gather and share large sets of computing resources, their aggregate energy consumption is an important problem to be addressed, given the economic and environmental impact of energy production and use. This is particularly true for file sharing peer-to-peer networks, considered the vast number of nodes participating to these systems. To address such problem, this paper proposes a peer-to-peer file sharing model organized in two logical layers. The lower layer is composed of a set of peers providing the files to be shared. The upper layer includes a set of super peers organized to form a distributed hash table, whose purpose is indexing files and peers, including their availability status. Following a sleep-and-wake approach, energy saving is achieved by letting peers cyclically switch between normal and sleep modes, where the time passed in normal mode by a peer depends on the number of files it provides. An energy-saving algorithm and an associated file search and retrieval strategy are proposed within this model. Simulation results show that more than 50% of energy can be saved using the proposed model and that more than 80% of file downloads start with no additional delay compared with a network in which all peers are always powered on. Copyright © 2014 John Wiley & Sons, Ltd.

With the advent of BitTorrent protocol, internet users can easily share their movie and music files. Although the BitTorrent provides a highly efficient mechanism, the illegal use of BitTorrent on unauthorized distributing of copyright- protected digital content is increasing rapidly. To prevent illegal distribution of copyrighted files from spreading over the internet, the file pieces split by torrent file must be detected as soon as possible. In this paper, we propose a method of audio recognition on a few pieces of audio files downloaded by BitTorrent client program. The audio identification algorithm is made up of beat tracking and chroma features. To identify audio in the piece, the beat-by-chroma representations obtained from the piece and original audio are cross-correlated and look for sharp peaks. The experimental results show probabilities of successfully identifying the pieces with different piece sizes and different number of pieces.

Because the Excel file is often edited, the data blocks stored in the hard disk which make up the Excel file is often not continuous. Discontinuous storage fragmented files, deleted or formatted, with the current data recovery software to restore the effect is very poor, often file recovery, but can not open or is garbled. Therefore it makes sense to study an effective recovery method to recover Excel fragmented files. Based on "Laura" file format, first introduced the Windows platform fragmented files, and then given an Excel file fragmentation recognition method, final based on debris identify given Excel file fragmentation restructuring.

Parallel file systems are experiencing more and more applications from various fields. Various applications have different I/O workload characteristics, which have diverse requirements on accessing storage resources. However, parallel file systems often adopt the "one-size-fits-all" solution, which fails to meet specific application needs and hinders the full exploitation of potential performance. This paper presents a framework to enable dynamic file I/O path selection with fine granularity at runtime. The framework adopts a file handle-rich scheme to allow file systems choose corresponding optimizations to serve I/O requests. Consistency control algorithms are proposed to ensure data consistency while changing optimizations at runtime. One case study on our prototype shows that choosing proper optimizations can improve the I/O performance for small files and large files by up to 40 and 64.4 %, respectively. Another case study shows that the data prefetch performance for real-world application traces can be improved by up to 193 % by selecting correct prefetch patterns. Simulations in large-scale environment also show that our method is scalable and both the memory consumption and the consistency control overhead can be negligible.

Versioning file systems are widely used in modern computer systems as they provide system recovery and old data access functions by retaining previous file system snapshots. However, existing versioning file systems do not perform well with the emerging PCM (phase-change memory) storage, because they are optimized for hard disks. Specifically, a large amount of additional writes incurred by maintaining snapshot degrades the performance of PCM seriously as write operations are the performance bottleneck of PCM. This paper presents a novel versioning file system, designed for PCM, that reduces the writing overhead of a snapshot significantly. Unlike existing versioning file systems that incur cascade writes up to the file system root, our scheme breaks the recursive update chain at the immediate parent level. The proposed file system is implemented on Linux 2.6 as a prototype. Measurement studies with various I/O benchmarks show that the proposed file system improves the I/O throughput by 144 percent on average, compared to ZFS, a representative versioning file system.

In this paper, we investigated the problem of copyright infringement in the peer-to-peer (P2P) environment, and proposed a system for piracy tracking. Even though the BitTorrent P2P software is a very efficient way to transfer files, it has become the main tool used to pirate copyrighted materials. Thus, we studied the bit torrent packet itself in order to detect the destination and source of the piracy. The torrent meta-file which has hash values of the pieces of the shared content is also analyzed to identify copyrighted material. As a result, we proposed a tracking system for copyright protection against BitTorrent piracy. Our system consists of hash database, feature database, protocol analyzer, torrent meta-file crawler, and copyright identifier. In this tracking system, the content is identified by three ways. The first way is to use the info hash in the hash DB. The second way is to use the piece hash value in the hash DB. Last way is to use the feature of the content that is able to identify if the content is slightly changed from original. If the proposed system is installed on the ISP network, it is able to protect the content from BitTorrent distribution.

Peer-to-Peer (P2P) file sharing attracts much attention due to its scalability and robustness. One important metric in measuring the performance of a P2P file sharing system is the amount of time required for all peers to get the files. We refer this time as the file distribution time. Researchers have proposed protocols to minimize the file distribution time under different situation. However, most works are based on the single-file scenario. On the other hand, there are studies showing that in a file sharing application, users may download multiple files at the same time. In this paper, we analyze the minimum time needed to distribute multiple files. We develop an explicit expression for the minimum amount of time needed to distribute multiple files in a heterogeneous P2P fluid model. Unlike the single-file scenario, we demonstrate that the theoretical lower bound in multi-file is not always achievable. With a comprehensive consideration of all the configurations, we elaborate how to partition the bandwidth capacities of both seeds and leechers for a particular file such that the finish time is optimal.

Based on the analysis of the structure of the 3DS file, this paper uses recursive method to collect content of 3DS file and designs the data structure to save them. Then, reading and drawing principle for 3DS file is studied by DirectX 11. Finally, the system for reading and drawing 3DS file is designed and implemented based on DirectX11 in VS2010, and achieves better results.

Constructing the tree type logic structure of a file and introducing the view mechanism of database system into the file system, the hierarchical structure of view files is created. Different files ( view files ) are provided to different users according to their levels or interests by mapping layer by layer with beginning of the minimum access unit (the logic block). It can solve the problem of the discrepancy of users views.

The hybrid file system is designed to optimize the latency of the response of File System I/Os and extend the capacity of the local file system to cloud by taking the advantage of Internet. Our hybrid file system is consist of SSD, HDD and Amazon S3 cloud file system. We store small files, directory tree and metadata of all the files in SSD, because SSD has a good performance for the response of small and random I/Os. HDD is good at responding big and sequential I/Os, so we use it just like a warehouse to store big files which are linked by the symbolic files in the SSD. We also extend the local file system to cloud in order to enlarge its capacity. In this paper we describe test data of our hybrid file system and also its design and implement details.

File carving is a recovery technique which does not consider file tables or other meta-data which is used to organize data on storage media. As files can be recovered based only on their content and/or structure, this technique is an indispensable task during digital investigations. The main contribution of this paper is the description of procedures that allow for successful content-based recovery of multimedia files from their fragments. So far many approaches for the recovery of digital images have been proposed. After a short discussion of relevant representatives in this domain the authors focus on the applicability of these approaches to the recovery of multimedia files.

Forensic Analyst encounters a mixed file fragments in the absence of File Table information i.e., files" metadata. File Carving is a process of reconstructing files from mixed file fragments without using files" metadata. File Carving is an interesting and challenging problem in digital forensics and Data Recovery. Recently there have been number of research papers in the area of File Carving. In this paper authors describe File Carving and present its Literature Survey. The implementation of Greedy Sequential Unique Path File Carving algorithm for 4-bit bitmap files, with the help of function prototypes using C language and brief explanation of these prototypes is explained. The experimental results are also shown. The experimental results show that the files can be reconstructed from their fragments without knowing files" metadata. However there are limitations of these methods as mentioned in section 5.

Distributed cryptographic file systems enable file sharing among their users and need the adoption of a key management scheme for the distribution of the cryptographic keys to authorized users according to their specific degree of trust. In this paper we describe the architecture of a basic secure file sharing facility relying on a multi-party threshold-based key-sharing scheme that can be overlaid on top of the existing stackable networked file systems, and discuss its application to the implementation of distributed cryptographic file systems. It provides flexible access control policies supporting multiple combination of roles and trust profiles. A proof of concept prototype implementation within the Linux operating system framework demonstrated its effectiveness in terms of performance and security robustness.

Nowadays, flash memory has drawn much attention of digital investigators, however most of them try to recover the content from logical aspect and few of them pay attention to how those files were created or modified. The deleted and edited contents of a file on the flash chips are commonly related to user behaviors which can be used as digital evidence. In this paper, a method using YAFFS2 metadata to recover files, reconstruct file system, and recover their previous history versions is proposed. The experimental results under Linux operating system show that the proposed method can correctly reconstruct file system, recover file and file traces from YAFFS2; and experiments conducted on physical images of Android phones show that our method can be applied to real scenarios.

Huge amount of data are stored in the network security environment and quantities of data in these fields tend to increase year on year. For this reason efficient File matching algorithms should be used which minimizes computer storage which results in minimum response time on searching process. There are many problems involved in file content matching technique. This technique works for finding the file content similarity. In this study, we propose the new algorithm called Adaptive File Comparison Technique, which converts the source and destination file contents into tokens and finally it makes a matrix format data called sparse matrix. The comparison technique compares the sparse matrices of source and destination files to find the content similarity and dissimilarity which leads to an efficient content matching technique rather than comparing the entire file content.

Organizing documents in the file system is one of the most tedious and thorny tasks for most computer users. Taxonomies based on hand made directory hierarchies still remain the only possible alternative for most small and medium enterprises, public administrations and individual users. However, both the limitations of the hierarchical organization of file systems and the difficulty of maintaining the coherence within the taxonomy have raised the need for more scalable and effective approaches. Desktop searching applications provide proprietary interfaces that enable content-based searching at the cost of having no control on the indexing and ranking of results. Semantic file systems, instead, leave users the freedom to manage the taxonomy according to their specific needs, but lose the standard file system features. In this paper we describe GFS (graph-based file system) a new hybrid file system that extends the standard hierarchical organization of files with semantic features. GFS allows the user to nest semantic spaces inside the directory hierarchy leaving unaltered system folders. Semantic spaces allow customized file tagging and leverage on browsing to guide file searching. Since GFS does not change the low-level interface to interact with file systems, users can continue to use their favorite file managers to interact with it. Moreover, no changes are required to integrate the semantic features in proprietary software.

File identification and analysis is an important process of computer forensics, since the process determines which subjects are necessary to be collected and analyzed as digital evidence. An efficient file classification aids in the file identification, especially in case of copyright infringement where we often have huge amounts of files. A lot of file classification methods have been proposed by far, but they have mostly focused on classifying malicious behaviors based on known information. In copyright infringement cases, we need a different approach since our subject includes not only malicious codes, but also vast number of normal files. In this paper, we propose an efficient file classification method that relies on undocumented information in the header of the PE format files. Out method is useful in copyright infringement cases, being applied to any sort of PE format executable file whether the file is malicious, packed, mutated, transformed, virtualized, obfuscated, or not.

One of the most highly regarded discrete event simulators used for network simulation is NS-2 (Network Simulator). NS-2 executes simulation scenarios producing various data of which trace files are considered the most beneficial for evaluating a simulation. In this paper we present the design considerations and implementation of a new tool that can be used both in TRAce FILe analysis and execution of simulations using NS-2. Although TRAFIL is primarily based on NS-2 trace files it can be extended to support a number of different other simulation trace file formats. It aims to make the execution of a great number of network simulations quicker, and the extraction of results from a large amount of data more flexible and productive. In order to accomplish the above tasks TRAFIL presents a novel way of interpreting, parsing, reading and eventually using NS-2 trace files. It introduces the notion of “metafiles” and “sub metafiles” throughout the procedures of trace file recognition and parsing, making the overall analysis operation substantially efficient and faster than alternative approaches. Metafiles and sub metafiles are used to encode NS-2 trace file structures enabling a more abstract approach to the trace file processing operation. Furthermore, TRAFIL facilitates the overall trace file analysis task by offering the opportunity to store each trace file as well as every Quality of Service (QoS) measurement produced for each trace file. Following the trace file recognition and processing operations, the information contained in a trace file is presented through a Graphical User Interface (GUI) offered by TRAFIL along with a variety of data, metrics and statistics related to simulation results. Finally, the tool offers the opportunity to execute custom Structured Query Language (SQL) queries to the local database and to completely automate the simulation procedure by enabling the user to execute NS-2 scripts as well as perform a simulation of a video transmission using the Evalvid-RA framework.

Objectives: Everyday millions of users are downloading files through internet. It usually takes more amount of time and often leads to internet traffic in order to satisfy various clients in distributed environment. At present, cloud provides vital storage solution for both enterprises and the individual users. Methods: The existing system uses dual direction file transfer technique to transfer a file from its either end of the direction in order to enhance the download speed. But this approach does not guarantee reliability and it also increases latency for file transfer if the servers are too far from each other. Moreover, this technique was invented for single client. If multiple servers want to download the same file it will increase the total time required for download. Findings: A novel file transfer tool using a message broker. Kafka has been proposed to transfer large sized file with enhanced reliability and security. Kafka is one of the latest message broker used by many enterprises. This technique reduces the time required for replicating the files on many servers and downloading into the target server. If there are multiple clients, Kafka makes use of its mirroring technique to make a copy of files on many servers and we can download it easily. Applications: File transfer using Kafka can be used by many companies for managing their log files and to transfer it quickly from one data center to another. It also ensures reliability and secure file transfer across the data centers in the world.

Transparent file encryption is an important means to protect file data security. However, for the traditional file encryption system based on file filter driver, constant cache cleaning is required to ensure the correctness of the data in the cache, which greatly reduces the efficiency of file operations. In the paper, a transparent file encryption system based on FUSE (File System in User Space) is proposed to overcome the shortcomings of the traditional transparent file encryption systems. To avoid cleaning the system file cache frequently, file redirection is adopted to transfer file operations to FUSE, which processes the file operations initiated by the trusted processes and non-trusted processes differently, with respect to cache. This system can not only be used for local file protection, but can also be applied to secure the files on cloud storages.

It has been observed that there has been a great interest in computing experiments which has been useful on shared nothing computers and commodity machines. We need multiple systems running in parallel working closely together towards the same goal. Frequently it has been experienced and observed that the distributed execution engine named MapReduce handles the primary input-output workload for such clusters. There are numerous distributed file systems around viz. Tinyfs, Glusterfs, Lustre, Cephfs, zfs and HDFS (Hadoop Distributed File System), we studied them and implemented MapReduce on few distributed file systems. It has been studied that distributed file systems (DFS) work very well on many small files but some do not generate expected output on large files. We implemented MapReduce algorithms onto three nodes of each distributed files systems for small and large files, and the analysis is been put forward in this paper. Even we came across the various implementation issues of various DFS, they have also been mentioned in this paper.

Most of the file systems today are based on desktop metaphor utilizing hierarchies of folders and sub-folders for information storage and retrieval. A folder is usually assumed to store the semantically related information in the form of files and sub-folders. However, there always exist related files somewhere else in the local file system hierarchy. Users often fail to find these disconnected files or discover the relationship among these. File systems maintain a plain set of metadata items about each file which could be utilized to discover some of the hidden relationships. However, this meta-information is very sparse as most of this metadata is rarely entered by the user. Moreover, files in a file system are mostly private and cannot get the benefit of collaborative tagging. Web of Data, on the other hand also contains information related to the file system objects which could be utilized for linking with information in file systems. The purpose of introducing Linked Data technology was to solve similar issues on the web. This article investigates the exploitation of Linked Data technology in file systems and provides a comprehensive review on the possible use of Linked Data Technology in file systems.

With the development of sensing technologies and popularization of intelligent mobile phone, crowd-sensing has entered the central stage of mobile computing. A huge amount of sensing data collected from all kinds of mobile intelligent terminals form lots of files with different sizes, which need to be effectively data stored method into a file system or a data warehouse. To solve these problems, a hybrid file storage method is designed in the paper, which can support hybrid storage for all kinds of files from sensors of crowd-sensing process by classifying files to big files and small files, and storing these two types of files by different storing and transferring strategies. In addition, a load balancing method of metadata servers based on a genetic algorithm is also proposed to make file storage more adaptive to dynamic changing loads. Finally, experiments have been conducted to verify the effectiveness of storage method for files sensing data designed in the paper.

Distributed file system has the characteristics of large capacity, good scalability and high reliability, which make it widely used in many areas involving large-scale data storage. It offers simplified, highly-available services for users to access data. However, due to the non-metadata design, the performance of traversal operation on large directories in those non-metadata distributed file systems is poor. With the increasing amount of files, it severely affects the user experience. In this paper, we present a directory prefetching mechanism on the client side to reduce directory traversal operation latency in non-metadata distributed file system. The mechanism, combined with the client's cache, adopts the directory access history to predict future access pattern and fetches the content of the directory without user intervention. Our goal is to reduce the overall access latency in the non-metadata distributed file system in order to better satisfy the user experience.

In the internet age, the security of data exchange becomes risky from some tapping attacks. One of the ways to maintain the data's security is by using steganography techniques. In this paper, steganography is combined with cryptography for security enhancement. The cover media is mp3 audio file. Since mp3 is a file that has been widely used in the Internet as a medium of entertainment, the secret message has been embedded to the homogenous frame of mp3 file. Before being embedded into the mp3 file, the secret message was encrypted with Advance Encryption Standard (AES) algorithm using a key that had been processed by Message Digest 5 (MD5) hash function. The secret message file type inserted could be in the form of text, image, audio, video, or compressed file. The result shows that the steganographic capacity does not depend on the size of mp3 but on the number of homogeneous frames owned by mp3 audio file. The result also shows that the method which combined steganography and cryptography, was successfully done without reducing the quality of used mp3 audio file.

HDFS is designed for storing large files, but it suffered performance penalty when storing large amount of small files such as the space occupied by the metadata cause high consumption of NameNode and low efficiency of file reading. Currently, there are many approaches implemented to solve the small file problem. In this paper we use additional hardware named SFS (Small File Server) between users and HDFS to solve the small file problem. The proposed approach includes a file merging algorithm based on temporal continuity, an index structure to retrieve small files and a prefetching mechanism to improve the performance of file reading and writing. The experimental results show that the proposed approach efficiently optimizes small files storing in HDFS with reducing the overload of NameNode and improving the performance of file accessing.

With the advent of the era of big data, XFS file system utilization rate is higher and higher because of its obvious advantages in terms of store huge amounts of data. But the corresponding detailed introduction of XFS file system is less. This paper studies the data storage mode in XFS file system, and analysis the change of disk after file being deleted. At last, by an example, we find a way to recover file by finding file inode through the file directory index then locating the blocks which belong to these inodes. Finally, we come to the conclusion that files which are deleted can be effectively restored when the inodes and the blocks belong to the file are not over written.

Parallel file systems are important infrastructures for both cloud and high performance computing. The performance of metadata operations is critical to achieve high scalability in parallel file systems. Nevertheless, traditional parallel file systems are lack of scalable metadata service. To alleviate these problems, some previous research distributes metadata to separated large-scale clusters and uses write-optimized techniques like log-structured merge tree (LSM-tree) to store metadata. However, LSM-tree design does not consider the features of solid state drive devices (SSD) which are widely deployed in modern parallel computing systems. The design of using LSM-trees to store metadata has not explored the potential benefits of SSD devices. In this paper, we present StageFS, which is a parallel file system optimized for SSD based clusters. StageFS stores both the metadata and small files in LSM-trees for fast indexing. For larger files, the file blocks are separately stored to reduce the write amplifications. In addition, the parallel I/O feature of SSD devices is used to improve the performance of accessing directories and large files. To avoid frequent small writes, StageFS uses buffering to better utilize the bandwidth of SSD devices. Experimental results show that StageFS provides better performance in metadata operations (up to 21.28x) and small file access (1.92x to two orders of magnitude) compared with Ceph and HDFS.

High performance computing (HPC) applications demand high I/O throughput from file systems to match their fast processing requirements. These HPC applications create large amounts of file meta-data that can overwhelm current single meta-data server file systems leading to performance bottlenecks. We address this problem with a multiple meta-data server (M-MDS) design using standard parallel NFS (pNFS). We utilize NFS directory referrals mechanism with hashing to efficiently distribute meta-data across a cluster of metadata servers. We show through a large scale setup in Amazon EC2 that our pNFS M-MDS can scale almost linearly and outperform Lustre CMD (Clustered Metadata) by up to 3 times in some of the file system meta-data operation benchmarks.

Although Mainline DHT is not an official BitTorrent protocol so far, it's widely used in many BitTorrent clients such as Utorrent, Vuze, BitComet and Xunlei. Despite there are millions of Mainline DHT users, only little research focus on Mainline DHT. In this paper, we present two kinds of measurement results based on crawling Mainline DHT. Firstly, we develop an effective crawler and crawl Mainline DHT over 10 days. We collected 10,149,649 magnet links and received over 264,846,482 get_peers requests from 57,065,499 unique peers. Based on the measurement results, we found that: (1) Russian and China play dominant roles in Mainline DHT which contribute 35% peers in it, and there are 5% Internet users use Mainline in Europe, (2) UTorrent is the most popular clients so far, half of peers use it. (3) We found that most clients use randomly chosen port numbers, only 8.9% use constant port. (4) Zona is a popular Russia peer to peer client, nine in ten top popular magnet links are unique in Zone, and only one is BitTorrent magnet link. (5) The peers' size of torrent almost follows a Zipf distribution. 11% torrents take up 95% peers of the total. 46% torrents only get one peer during our experiment. Secondly, we study two ways to convert magnet link to torrent file: Converting by five popular torrent storage services and converting by Extension protocol. We compare the performance of them in terms of converting rate, converting time, advantages and disadvantages and discover that: (1) The converting rate of torrent storage services and Extension protocol are 46.9% and 22.7%. (2) The average successful converting time of Extension protocol is 75s, and if a magnet spends more than 150s to convert but still couldn't get good peers, it tends to convert unsuccessfully at present. The best scheme is stopping for a while, and waiting good peers on line. (3) NAT is the main reason caused the converting or downloading failure, there are 32.5% peers stay in NAT or firewall. The best solution of NAT user is the popularized of IPv6 or Tunneling. Understanding DHT in depth can improve the performance of current DHT system.

Sensitive information leakage in files is one of the biggest concerns in people's life. Considering this situation, people have designed lots of tools to protect sensitive files. However, these specialized tools are not applicable when the user amount is small. This paper proposes a novel light-weight mechanism that provides efficient and fast cryptographic-based file protection to meet the need of small organizations. The highlights are its ability to provide protection for all types of files, and the usage of an efficient access control approach for file sharing. Operations in the directory are monitored in real time and files are protected automatically. Moreover, encryption keys are generated and well preserved by a hierarchical key generation algorithm. For further estimation, we conducted experiments and analyzed the security and performance of our mechanism as well as other file protection software, and then we made a comparison among them. Experimental results show that our mechanism performs well both in security and performance and outperforms those other file protection software when applied in small organizations.

MapReduce is a popular computing model for parallel data processing on large-scale datasets, which can vary from gigabytes to terabytes and petabytes. Though Hadoop MapReduce normally uses Hadoop Distributed File System (HDFS) local file system, it can be configured to use a remote file system. Then, an interesting question is raised: for a given application, which is the best running platform among the different combinations of scale-up and scale-out Hadoop with remote and local file systems. However, there has been no previous research on how different types of applications (e.g., CPU-intensive, data-intensive) with different characteristics (e.g., input data size) can benefit from the different platforms. Thus, in this paper, we conduct a comprehensive performance measurement of different applications on scale-up and scaleout clusters configured with HDFS and a remote file system (i.e., OFS), respectively. We identify and study how different job characteristics (e.g., input data size, the number of file reads/writes, and the amount of computations) affect the performance of different applications on the different platforms. This study is expected to provide a guidance for users to choose the best platform to run different applications with different characteristics in the environment that provides both remote and local storage, such as HPC cluster.

A file can be called an atomic information element that is found all over in computer world just like dust particles in real world. Files are universally used, ranging from personal data on desktop computers, to organizational data shared over local networks and the data that exist on the web. That's why files caught the attention of ours and lead us to study the life of a file from Cradle to Grave or Eternity. In our study we investigate: the birth of a file; sources of file's birth; the tsunami of files; the good, the bad and the ugly files; file aging; file states transition; file lost and the quest of finding a file; a journey in past and future; the death of a file; file resurrection and the gods of clouds.

The world keeps contributing to the increase in data everyday drastically. Scientific applications, weather forecasting, researches, hospitals, military services are few such major contributors. As the amount of data increases, the need to provide efficient, easy to use solutions has become one of the main issues for these type of computations. The best solution to this issue is the use of Distributed File Systems(DFS). Some existing Distributed File Systems are too complex to deploy and maintain, although they are extremely scalable and cheap since they can be entirely built out of commodity Operating System(OS) and hardware. GlusterFS solves this problem. Gluster File System(GFS) is open source and is capable of scaling to several petabytes(upto 72 brontobytes) and handling thousands of clients. It is based on a stackable user space design and can deliver exceptional performance for diverse workloads. GlusterFS is an easy to use clustered file system that meets enterprise-level requirements. It is written in user space which uses FUSE(Filesystem in user space) to hook itself with the VFS layer. It takes a layered approach to the file system, where features are added/removed as per the requirement. Enforcer is a major component in Gluster which finds its extensive use in production. Enforcer is concerned more about limits. It helps to restrict the usage(both in terms of size and count) at directory or volume level in the file system. In this paper, we discuss the internal working of the translators and key architectural components of GlusterFS. Along with that, one can gain a deep insight about the enforcement techniques, few challenges, undergoing researches and projects of GlusterFS

The importance of security is increasing in a current network system. We have found a big security weakness at the file integration when the people download or upload a file and propose a novel solution how to ensure the security of a file. In particular, hash value can be applied to ensure a file due to a speed and architecture of file transfer. Hash server stores all the hash values which are updated by file provider and client can use these values to entrust file when it downloads. FileShader detects to file changes correctly, and we observed that it did not show big performance degradation. We expect FileShader can be applied current network systems practically, and it can increase a security level of all internet users.

A storage system in cloud is well thought-out as a very big scale storage system that has independent storage servers. The service that cloud storage provides is, that can store user's data from remote through network and other authenticated users can access the data easily. Hadoop distributed file system is used to store large files consistently and to retrieve those files at very high bandwidth to user applications. Hadoop splits the files into large blocks and distributes them amongst the nodes in the cluster. When we retrieve data from the cloud, it is very important that the computation and communication overhead should be reduced. To reduce the communication overhead the server should send only the top-n files based on the keyword when the user asks for the data files. Since the owner need not maintain the copy of the files, it is all the more necessary to make check on the files available and also check the originality of the files stored in the server periodically. In HDFS the computation is done in parallel so that the execution time is drastically reduced. In the proposed system for retrieving top-n files we use Hadoop Distributed File System, so that the search time and the communication overhead is greatly reduced.

Recovering deleted files play an important role in a digital forensic investigation. When a file is deleted, only pointers that link file's metadata to its content are deleted and metadata entry is marked as deleted. As long as data is not overwritten or wiped, deleted data will remain in unallocated space. One of the methods that can be used to recover these deleted files is file carving. File carving reconstructs files only based on their content unlike traditional data recovery methods that use metadata that points to the content. It is mainly done using headers and footers of file types. One of the primary challenges in file carving is to recover deleted files that were fragmented. When a file is fragmented, carving only based on header and footer will produce corrupted file. So this paper discusses a method for carving fragmented document and image files from a FAT32 formatted USB drive.

Using an alternative approach to traditional file hashing, digital forensic investigators can hash individually sampled subject drives on sector boundaries and then check these hashes against a prebuilt database, making it possible to process raw media without reference to the underlying file system.