Search

Virtual machines (VMs) have become the predominant paradigm for deploying applications, and flash-based caches are widely deployed for VMs to improve their performance. However, due to the semantics of journaling in the VMs' file system are transparent to flash caches, traditional flash caching with journaling induces inefficient logging I/O traffic to the shared storage server and duplicated caching in flash, which not only underutilizes flash caches but also aggravates their wearout. This paper presents JCache, a journaling-aware flash caching solution to address these problems. First, JCache proposes the virtual journal device design to receive and deliver the journaling semantics in the VMs to the cache manager so as to optimize the VM's journaling. Second, JCache proposes a cache-only logging mechanism, which transparently uses the persistent flash caches as the journal area and does cache-only logging to safely eliminate the logging I/O traffic to the shared storage server. Third, JCache proposes the logical caching mechanism, which eliminates the duplicated flash caching induced by logging I/Os and in-place updates to mitigate the wearout of flash caches. Evaluations show that JCache improves the application performance by up to 11.4× and reduces the flash cache writes by up to 42% compared to traditional flash caching solutions.

Public key encryption with keyword search (PEKS) allows a cloud server to retrieve particular ciphertexts without leaking the contents of the searched ciphertexts. This kind of cryptographic primitive gives users a special way to retrieve the encrypted documents they need while preserving privacy. Nevertheless, most existing PEKS schemes only offer single-keyword search or conjunctive-keyword searcha. The poorly expressive ability and constantly inaccurate search results make them hard to meet users' requirements. Although several expressive PEKS (EPEKS) schemes were proposed, they entail high computation and communication costs. An ideal EPEKS scheme should enable fast and accurate ciphertext retrieval, while lowering the storage server's load and reducing the amount of communication data. Drawing on the strongly expressive ability of key-policy attribute-based encryption (KP-ABE), we propose a generic construction of EPEKS from KP-ABE. We demonstrate that the derived EPEKS scheme is secure under the chosen keyword attack if the implicit KP-ABE scheme fulfills the anonymity under the chosen plaintext attack. Furthermore, we present a concrete EPEKS scheme over the prime-order groups. The comparison and experimental results indicate that our scheme is more efficient than the existing EPEKS schemes.

The security and privacy of electronic health records (EHRs) have received considerable attention from healthcare workers and researchers. To ensure security, various encryption and decryption schemes as well as key management protocols have been developed. However, owing to sharing and scalability issues, additional security technologies have been proposed. Nonetheless, these technologies cause other problems, such as efficiency issues. Blockchain-based EHR management systems have been proposed to overcome computational overhead. However, because most blockchain systems are installed by outsourcing companies, EHRs may be leaked to the company. Hence, we herein propose a blockchain-based EHR management scheme with proxy re-encryption. In this scheme, we set a proxy server that re-encrypts the ciphertext between file servers, thereby solving EHR sharing issues. Furthermore, because the server is separated from the blockchain system, the outsourcing company cannot manipulate the server or access the records. In addition, the blockchain assists in access control by using smart contracts, thereby enabling secure and efficient EHR sharing. By performing security analysis, we prove that our proposed scheme solves the aforementioned security problems. In addition, we experimentally demonstrate the efficient operation of the proposed system.

Today more and more enterprises and individuals are outsourcing their data to cloud storage system. Data deduplication is one of the important technologies to reduce the storage cost of cloud storage system. In a cloud storage system with deduplication technology, the client can outsource the data files to the cloud storage server and pay for them. Fair payment is one of key issues in the cloud deduplication storage system. At present, a variety of secure deduplication encryption schemes have been designed to protect the privacy of client data. However, most existing fair payment solutions use traditional electronic cash systems to generate payment tokens, which requires a trusted authority to prevent double-spending. Trusted authorities will become bottlenecks in the payment system. Faced with this problem, in this paper, we propose a new decentralized fair payment protocol for cloud deduplication storage system by utilizing ethereum blockchain technology. The new protocol takes advantage of the decentralization of blockchain technology, allowing direct transactions without the participation of trusted third parties. In the new protocol, if a malicious situation occurs, the system can guarantee fair payment by pre-storing penalty money in the smart contract. Safety analysis and experimental analysis show that our new protocol is feasible.

The Internet of Things (IoT) is also known as the Internet of everything. As an important part of the new generation of intelligent information technology, the IoT has attracted the attention both of researchers and engineers all over the world. Considering the limited capacity of smart products, the IoT mainly uses cloud computing to expand computing and storage resources. The massive data collected by the sensor are stored in the cloud storage server, also the cloud vulnerability will directly threaten the security and reliability of the IoT. In order to ensure data integrity and availability in the cloud and IoT storage system, users need to verify the integrity of remote data. However, the existing remote data integrity verification schemes are mostly based on the RSA and BLS signature mechanisms. The RSA-based scheme has too much computational overhead. The BLS signature-based scheme needs to adopt a specific hash function, and the batch signature efficiency in the big data environment is low. Therefore, for the computational overhead and signature efficiency issues of these two signature mechanisms, we propose a scheme of data integrity verification based on a short signature algorithm (ZSS signature), which supports privacy protection and public auditing by introducing a trusted third party (TPA). The computational overhead is effectively reduced by reducing hash function overhead in the signature process. Under the assumption of CDH difficult problem, it can resist adaptive chosen-message attacks. The analysis shows that the scheme has a higher efficiency and safety.

In cloud computing, a shared storage server, which provides a network-attached storage device, is usually used for centralized data management. However, when multiple virtual machines (VMs) concurrently access the storage server through the network, the performance of each VM may decrease due to limited bandwidth. To address this issue, a flash-based storage device such as a solid state drive (SSD) is often employed as a cache in the host server. This host-side flash cache saves remote data, which are frequently accessed by the VM, locally in the cache. However, frequent VM migration in the data center can weaken the effectiveness of a host-side flash cache as the migrated VM needs to warm up its flash cache again on the destination machine. This study proposes Cachemior, Firepan, and FirepanIF for rapid flash-cache migration in cloud computing. Cachemior warms up the flash cache with a data preloading approach using the shared storage server after VM migration. However, it does not achieve a satisfactory level of performance. Firepan and FirepanIF use the source node&rsquo, s flash cache as the data source for flash cache warm-up. They can migrate the flash-cache more quickly than conventional methods as they can avoid storage and network congestion on the shared storage server. Firepan incurs downtime of the VM during flash cache migration for data consistency. FirepanIF minimizes the VM downtime with the invalidation filter, which traces the I/O activity of the migrated VM during flash cache migration in order to invalidate inconsistent cache blocks. We implement and evaluate the three flash cache migration techniques in a realistic virtualized environment. FirepanIF demonstrates that it can improve the performance of the I/O workload by up to 21.87% compared to conventional methods.

Oblivious Random Access Machine (ORAM) constructions can be used to hide a client’s access pattern from a trusted but curious storage server. The privacy provided comes at the cost of increasing communication overhead, storage overhead, and computation overhead of the system. Recursive Matrixbased ORAM (RM-ORAM) is a new ORAM construction which is designed for constrained storage space devices. RM-ORAM significantly reduces the client storage usage by using recursion while the computational and bandwidth overhead are slightly increased as a trade-off. However, it can achieve better overall asymptotic performance compared with other existing ORAM schemes, e.g. recursive Path ORAM. In this paper, we present the construction and its theoretical analysis. In addition, we present how to select the appropriate number of data blocks which are being downloaded per level of recursion and the appropriate size of reserved space on the client. We provide theoretical security and performance analysis, as well as experimental results to illustrate how RM-ORAM satisfies security requirements and provides improved performance compared to other ORAM schemes.

The rapid development of cloud storage technology paired with the prevalence of smartphone usage presents wide-ranging challenges for digital forensics practitioners. Data are more easily uploaded and shared between multiple devices and across multiple platforms. So, the process has increased the opportunities for criminality. Criminality undertaken in cloud computing can be directly seen on logs stored on the cloud storage server, which records user activity. However, because of user privacy protection, these logs cannot be easily used as evidence in court. This issue emphasizes the need for a reliable means of identifying, acquiring, and preserving evidential data from the client-side. This study identifies the data artifacts of a user accessing Dropbox via smartphone (Android Lollipop and Android Nougat). The data are from performing several common activities such as installing, signing up, uploading, downloading, sharing, and others. About 14 artifacts are identified by documenting the Dropbox client database changing contents as these activities are carried out. This study increases knowledge of the artifacts that are leftover by Dropbox client on Android smartphones. The results propose this comparing and analyzing method can be used by digital forensics investigators in carrying out investigations and cyberlaw practitioners as guidance in criminal cases.

The traditional flow of coastal ocean model data is from High-Performance Computing (HPC) centers to the local desktop, or to a file server where just the needed data can be extracted via services such as OPeNDAP. Analysis and visualization are then conducted using local hardware and software. This requires moving large amounts of data across the internet as well as acquiring and maintaining local hardware, software, and support personnel. Further, as data sets increase in size, the traditional workflow may not be scalable. Alternatively, recent advances make it possible to move data from HPC to the Cloud and perform interactive, scalable, data-proximate analysis and visualization, with simply a web browser user interface. We use the framework advanced by the NSF-funded Pangeo project, a free, open-source Python system which provides multi-user login via JupyterHub and parallel analysis via Dask, both running in Docker containers orchestrated by Kubernetes. Data are stored in the Zarr format, a Cloud-friendly n-dimensional array format that allows performant extraction of data by anyone without relying on data services like OPeNDAP. Interactive visual exploration of data on complex, large model grids is made possible by new tools in the Python PyViz ecosystem, which can render maps at screen resolution, dynamically updating on pan and zoom operations. Two examples are given: (1) Calculating the maximum water level at each grid cell from a 53-GB, 720-time-step, 9-million-node triangular mesh ADCIRC simulation of Hurricane Ike, (2) Creating a dashboard for visualizing data from a curvilinear orthogonal COAWST/ROMS forecast model.

In a scenario where user files are stored in a network shared volume, a single computer infected by ransomware could encrypt the whole set of shared files, with a large impact on user productivity. On the other hand, medium and large companies maintain hardware or software probes that monitor the traffic in critical network links, in order to evaluate service performance, detect security breaches, account for network or service usage, etc. In this paper we suggest using the monitoring capabilities in one of these tools in order to keep a trace of the traffic between the users and the file server. Once the ransomware is detected, the lost files can be recovered from the traffic trace. This includes any user modifications posterior to the last snapshot of periodic backups. The paper explains the problems faced by the monitoring tool, which is neither the client nor the server of the file sharing operations. It also describes the data structures in order to process the actions of users that could be simultaneously working on the same file. A proof of concept software implementation was capable of successfully recovering the files encrypted by 18 different ransomware families. This work was supported by Spanish MINECO through project PIT (TEC2015-69417-C2-2-R).

Implementation of practicum courses in Department of Computer Engineering Diponegoro University has some drawbacks, one of them is a lot of lab assistant and the practitioner experiencing difficulties in terms of archiving. One solution to solve the problem is implementing a shared file storage system that is easy and can be accessed by both practitioners or lab assistants. The purpose of this research is to build a cloud-based storage systems that are reliable to preventing crash damage hardware and high availability. The purpose of this research is achieved by designing the appropriate methodology. The result of this research is a storage system that is on the server side by using virtualization and data replication (DRBD) as a storage method. The system is composed of two physical servers and one virtual server. Physical servers are using Proxmox VE as operating system and virtual server is using Ubuntu Server as operating system. OwnCloud applications and files are stored in the virtual server. File storage system has several major functions, which are: upload, download, user management, remove, and restore. The functions are executed through web pages, desktop application and Android application.

The amount of forensic data collected by governments is increasing rapidly. Public cloud providers could store this data for a lower price than government owned infrastructure ever could, but for privacy reasons such data must be encrypted before it is sent to a public cloud. In this paper we present our SecureCloud system which provides a secure, efficient and seamless way to store forensic data in cheap but untrusted cloud environments. The forensic tools, however, are still executed on premises of the government agency, because they need cleartext data. We analyze the performance of our system using Autopsy to show that the forensic tool runs only 19 percent slower when reading encrypted data from SecureCloud compared with reading the same data from a local file server.

International audience; This paper presents an initiative data prefetching scheme on the storage servers in distributed file systems for cloud computing. In this prefetching technique, the client machines are not substantially involved in the process of data prefetching, but the storage servers can directly prefetch the data after analyzing the history of disk I/O access events, and then send the prefetched data to the relevant client machines proactively. To put this technique to work, the information about client nodes is piggybacked onto the real client I/O requests, and then forwarded to the relevant storage server. Next, two prediction algorithms have been proposed to forecast future block access operations for directing what data should be fetched on storage servers in advance. Finally, the prefetched data can be pushed to the relevant client machine from the storage server. Through a series of evaluation experiments with a collection of application benchmarks, we have demonstrated that our presented initiative prefetching technique can benefit distributed file systems for cloud environments to achieve better I/O performance. In particular, configuration-limited client machines in the cloud are not responsible for predicting I/O access operations, which can definitely contribute to preferable system performance on them

Cloud Computing is a service by which clients can outsource their data to reduce the burdens of data storage and maintenance. However, cloud providers are untrustworthy, which therefore introduce several security challenges: data integrity, data availability and data confidentiality. Since integrity and availability are prerequisite conditions for the existence of a system, researchers focus on these before confidentiality. To ensure integrity and availability, network coding-based POR (Proof of Retrievability) schemes have been proposed to enable the client to check whether the distributed data stored in the cloud servers is intact or not. However, most existing schemes have not achieved exact-repair feature and efficiency. In this paper, we propose a new POR scheme named SW-POR (Slepian-Wolfbased Proof of Retrievability) to address these gaps. Our scheme supports exact repair in which the repaired data is the same as the original corrupted data. Our scheme also can reduce the storage, computation, communication overheads and the size of a coded block. We proposed SW-POR based on Slepian-Wolf data compression code. We prove our security using entropy theory and implement SW-POR to show the costs.

The purpose of this research is to implement Server Web Hosting based Linux Ubuntu in Computer System/network of Elementary School 15 (SDN 15) Pangkalpinang, because there is not available of Computer Server Web Hosting as storage media of website based application and school website at Elementary School 15 Pangkalpinang, makes the school wants to create a unit of Server Web Hosting. The research methods of this thesis is analitic method and design method where at analitic method , writer analyzes the needs and with the analysis writer designs the steps to be done. Then , kind of this research is applied research, it means to give solution of a certain problem practically and to support development activities or implementation of a system , such as database, programming language, network concept. In order that project implementation is on time , so project model is used. The result of this is Server Web Hosting which can give services like DNS Server, FTP Server, File Server, Mail Server that can be accessed by clients through certain links. The conclusion, after server is implementatied can facilitate every one to access any informations from Elementary School 15 fast and easily anytime.

Traditional standalone embedded system is limited in their functionality, flexibility, and scalability. Fog computing platform, characterized by pushing the cloud services to the network edge, is a promising solution to support and strengthen traditional embedded system. Resource management is always a critical issue to the system performance. In this paper, we consider a fog computing supported software-defined embedded system, where task images lay in the storage server while computations can be conducted on either embedded device or a computation server. It is significant to design an efficient task scheduling and resource management strategy with minimized task completion time for promoting the user experience. To this end, three issues are investigated in this paper: 1) how to balance the workload on a client device and computation servers, i.e., task scheduling, 2) how to place task images on storage servers, i.e., resource management, and 3) how to balance the I/O interrupt requests among the storage servers. They are jointly considered and formulated as a mixed-integer nonlinear programming problem. To deal with its high computation complexity, a computation-efficient solution is proposed based on our formulation and validated by extensive simulation based studies.

KINematics And RIgidity (KINARI) is an on-going project for in silico flexibility analysis of proteins. The new version of the software, Kinari-2, extends the functionality of our free web server KinariWeb, incorporates advanced web technologies, emphasizes the reproducibility of its experiments, and makes substantially improved tools available to the user. It is designed specifically for large scale experiments, in particular, for (a) very large molecules, including bioassemblies with high degree of symmetry such as viruses and crystals, (b) large collections of related biomolecules, such as those obtained through simulated dilutions, mutations, or conformational changes from various types of dynamics simulations, and (c) is intended to work as seemlessly as possible on the large, idiosyncratic, publicly available repository of biomolecules, the Protein Data Bank. We describe the system design, along with the main data processing, computational, mathematical, and validation challenges underlying this phase of the KINARI project.

An emerging trend in developing large and complex applications on today’s high-performance computers is to couple independent components into a comprehensive application. The components may employ the global file system to exchange their data when executing the application. In order to reduce the time required for input/output (I/O) data exchange and data transfer in the coupled systems or other applications, this article proposes a dynamic process migration mechanism on the basis of block access pattern similarity for utilizing the local file cache to exchange the data. We first introduce the scheme of the block access counting diagram to profile the process access pattern during a time period on the storage server. Next, we propose an algorithm that compares the access patterns of processes running on different computing nodes. Last, processes are migrated in order to group processes with similar access patterns. Consequently, the processes on the computing node can exchange their data by accessing the local file cache, instead of the global file system. The experimental results show that the proposed process migration mechanism can reduce the execution time required by the application because of the shorter I/O time, as well as yield attractive I/O throughput. In summary, this dynamic process migration technique can work fairly well for distributed applications whose data dependency rely on distributed file systems.

Today's cloud storage services must offer storage reliability and fast data retrieval for large amount of data without sacrificing storage cost. We present SEARS, a cloud-based storage system which integrates erasure coding and data deduplication to support efficient and reliable data storage with fast user response time. With proper association of data to storage server clusters, SEARS provides flexible mixing of different configurations, suitable for real-time and archival applications. Our prototype implementation of SEARS over Amazon EC2 shows that it outperforms existing storage systems in storage efficiency and file retrieval time. For 3 MB files, SEARS delivers retrieval time of $2.5$ s compared to $7$ s with existing systems., 4 pages, IEEE LCN 2015

International audience; In Cloud storage of multiple CPU cores, many Mapreduce applications may run in parallel on each compute node and collocate with local Disks storage. These Disks storage are shared by multiple applications that use full CPU power of the node. Each application tends to issue contigu- ous I/O requests in parallel to the same Disk; however if large number of Mapreduce tasks enters the I/O phase at the same time, the requests from the same task may be interrupted by the requests of other tasks. Then, the I/O nodes receive these requests as non-contiguous way under I/O con- tention. This interleaved access pattern causes performance degradation for Mapreduce application, this is particularly important when writing intermediate files by multiple tasks in parallel to the shared Disk stor- age. In order to overcome this problem, we have proposed approach for optimizing write access for Mapreduce application. The contributions of this paper are: 1) analyze the open issues on scheduling access request of Mapreduce workload; 2) propose framework for scheduling and predicting I/O request of Mapreduce application; 3) describe each role of compo- nent that intervenes in the scheduling theses I/O request on Block-level of storage server to provide contiguous access

In this demonstration we present a platform that encompasses all of the components required to realistically evaluate the performance of Dynamic Adaptive Streaming over HTTP (DASH) over a real-time NS-3 simulated network. Our platform consists of a network-attached storage server with DASH video clips and a simulated LTE network which utilises the NS-3 LTE module provided by the LENA project. We stream to clients running an open-source player with a choice of adaptation algorithms. By providing a user interface that offers user parametrisation to modify both client and LTE settings, we can view the evaluated results of real-time interactions between the network and the clients. Of special interest is that our platform streams actual video clips to real video clients in real-time over a simulated LTE network, allowing reproducible experiments and easy modification of LTE and client parameters. The demonstration showcases how changes in LTE network settings (fading model, scheduler, client distance from eNB, etc.), as well as video-related decisions at the clients (streaming algorithm, quality selection, clip selection, etc.), can impact on the delivery and achievable quality.

As we move towards the Exascale era of supercomputing, node-level failures are becoming more common-place; frequent checkpointing is currently used to recover from such failures in long-running science applications. While compute performance has steadily improved year-on-year, parallel I/O performance has stalled, meaning checkpointing is fast becoming a bottleneck to performance. Using current file systems in the most efficient way possible will alleviate some of these issues and will help prepare developers and system designers for Exascale; unfortunately, many domain-scientists simply submit their jobs with the default file system configuration.\ud \ud In this paper, we analyse previous work on finding optimality on Lustre file systems, demonstrating that by exposing parallelism in the parallel file system, performance can be improved by up to 49x. However, we demonstrate that on systems where many applications are competing for a finite number of object storage targets (OSTs), competing tasks may reduce optimal performance considerably. We show that reducing each job's request for OSTs by 40% decreases performance by only 13%, while increasing the availability and quality of service of the file system. Further, we present a series of metrics designed to analyse and explain the effects of contention on parallel file systems. Finally, we re-evaluate our previous work with the Parallel Log-structured File System (PLFS), comparing it to Lustre at various scales. We show that PLFS may perform better than Lustre in particular configurations, but that at large scale PLFS becomes a bottleneck to performance. We extend the metrics proposed in this paper to explain these performance deficiencies that exist in PLFS, demonstrating that the software creates high levels of self-contention at scale.

Cloud computing has pioneered the emerging world by manifesting itself as a service through internet and facilitates third party infrastructure and applications. While customers have no visibility on how their data is stored on service provider’s premises, it offers greater benefits in lowering infrastructure costs and delivering more flexibility and simplicity in managing private data. The opportunity to use cloud services on pay-per-use basis provides comfort for private data owners in managing costs and data. With the pervasive usage of internet, the focus has now shifted towards effective data utilization on the cloud without compromising security concerns. In the pursuit of increasing data utilization on public cloud storage, the key is to make effective data access through several fuzzy searching techniques. In this paper, we have discussed the existing fuzzy searching techniques and focused on reducing the searching time on the cloud storage server for effective data utilization. Our proposedAsymmetric Classifier Multikeyword Fuzzy Searchmethod provides classifier search server that creates universal keyword classifier for the multiple keyword request which greatly reduces the searching time by learning the search path pattern for all the keywords in the fuzzy keyword set. The objective of using BTree fuzzy searchable index is to resolve typos and representation inconsistencies and also to facilitate effective data utilization.

The efficient distribution of stored information has become a major concern in the Internet. Since the Web workload characteristics shows that more than 60 % of network traffic is caused by image documents, how to efficiently distribute image documents from servers to end clients is an important issue. The proxy cache is an efficient solution to reduce network traffic. It has been shown that an image caching method (graceful caching) based on a hierarchical coding format showed better performance than conventional caching schemes. However, as the capacity of the cache is limited, how to efficiently allocate the cache memory to achieve a minimum expected delay time is still a problem to be resolved. This paper presents an integrated caching algorithm to deal with the above problem in the Internet. By analyzing the Web request distribution of graceful caching, both replacement and pre-fetching algorithms are proposed. We also show that our proposal can be carried out based on information readily available in the proxy server and it flexibly adapts its parameters to the hit rates and access pattern of users' requesting documents in the graceful caching. We finally verify the performance of this algorithm by simulations.

International audience; Recent technological advances have given rise to the popularity and success of cloud storage. However, the prospect of outsourcing an increasing amount of data to a third party and the abstract nature of the cloud foster the proliferation of security and privacy challenges, namely, the remote data possession checking. This paper addresses this critical security concern, when storing sensitive data in a cloud storage service, and the need for users to trust commercial cloud providers. It proposes a deterministic Proof of Data Possession (PDP) scheme based on Interactive Proof System(IPS) and an original usage of the GPS scheme. Our approach has several advantages. First, it supports public verifiability which releases data owners from the burden of a periodical verification. Second, it provides constant communication complexity, where the exchanged messages between the storage server and the client are composed of constant number of group elements. Third, our solution is efficient and provably secure, as it is resistant to the fraudulence of the prover and the leakage of verified data

The System Information Laboratory of Information Technology Department Polytechnic State of Padang has 30 units computer as education facilities to support learning process. All of computers used at same time in a learning section. This case causing trouble to monitoring each students activities. In order to get the solution for the lecturer, the writer then construct a server by using Linux operation system and client by using windows system operation in which Samba File Server is needed. By using this samba, the lecturer will be able to share the data and will be able to use the server as data storage media. Besides that, the writer will also use VNC (Virtual network connection) to simplify the process of monitoring and supervising client working system. Based on the result gotten after the writer done some experiment, it can be concluded that Samba File Server can also be used after some configuration is applied on certain files. Moreover, the writer also conclude that VNC can control the entire of the client. The writer suggests that Samba File server which will be used is the latest version one which has more feature than the previous one, it is suggested that the configuration of VNC is applied on Ubuntu Linux since the service is available. Kata Kunci : Samba File Server, VNC, Ubuntu installation

In many cases robots are connected wirelessly with a file server and often with one another, either directly, or via the file server. The network connections form a subnet where the router has the static IP address visible to the outside world and the server along with the robots form a subnet with local IP addresses. Often however, each robot has its own static IP address. In addition, each robot has a NIC card and a unique NIC address, as well as other hardware identifiers depending on the functionality and complexity of the robot. The non-electronic part of the robot hardware usually represents mature technology that has been understood for a long time. The electronic hardware has evolved to the point that the embedded software can provide the needed intelligence for the robot to perform sophisticated tasks previously performed by one or more human beings. However, in previous research emphasis has been placed on the tasks performed by the robots, neglecting any security issues or liabilities that may arise due to lack of security. In this paper, we provide an algorithm for secure key management, and secure communication in an insecure wireless and noisy environment in which the robots operate.

File systems are designed to control how files are stored and retrieved. Without knowing the context and semantics of file contents, file systems often contain duplicate copies and result in redundant consumptions of storage space and network bandwidth. It has been a complex and challenging issue for enterprises to seek deduplication technologies to reduce cost and increase the storage efficiency. To solve such problem, researchers proposed in-line or offline solutions for primary storages or backup systems at the subfile or whole-file level. Some of the technologies are used for file servers and database systems. Fewer studies focus on the cloud file system deduplication technologies at the application level, especially for the Hadoop distributed file system. It is the goal of this paper to design a file deduplication framework on Hadoop distributed file system for cloud application developers. The architecture, interface, and implementation experiences are also shared in this paper.

International audience; —Data grows at the impressive rate of 50% per year, and 75% of the digital world is a copy 1 ! Although keeping multiple copies of data is necessary to guarantee their availability and long term durability, in many situations the amount of data redundancy is immoderate. By keeping a single copy of repeated data, data deduplication is considered as one of the most promising solutions to reduce the storage costs, and improve users experience by saving network bandwidth and reducing backup time. However, this solution must now solve many security issues to be completely satisfying. In this paper we target the attacks from malicious clients that are based on the manipulation of data identifiers and those based on backup time and network traffic observation. We present a deduplication scheme mixing an intra-and an inter-user deduplication in order to build a storage system that is secure against the aforementioned type of attacks by controlling the correspondence between files and their identifiers, and making the inter-user deduplication unnoticeable to clients using deduplication proxies. Our method provides global storage space savings, per-client bandwidth network savings between clients and deduplication proxies, and global network bandwidth savings between deduplication proxies and the storage server. The evaluation of our solution compared to a classic system shows that the overhead introduced by our scheme is mostly due to data encryption which is necessary to ensure confidentiality.

International audience; Virtualization has become a prominent tool in data centers and is extensively leveraged in cloud environments: it enables multiple virtual machines (VMs) -- with multiple operating systems and applications -- to run within a physical server. However, virtualization introduces the challenging issue of preserving the high disk utilization (i.e., reducing the seek delay and rotation overhead) when allocating disk resources to VMs. Exploiting spatial locality, a key technique for improving disk utilization and performance, faces additional challenges in the virtualized cloud because of the transparency feature of virtualization (hypervisors do not have the information about the access patterns of applications running within each VM). To this end, this paper contributes a novel disk I/O scheduling framework, named Pregather, to improve disk I/O efficiency through exposure and exploitation of the special spatial locality in the virtualized environment (regional and sub-regional spatial locality corresponds to the virtual disk space and applications' access patterns, respectively), thereby improving the performance of disk-intensive applications without harming the transparency feature of virtualization (without a priori knowledge of the applications' access patterns). The key idea behind Pregather is to implement an intelligent model to predict the access regularity of sub-regional spatial locality for each VM. We implement the Pregather disk scheduling framework and perform extensive experiments that involve multiple simultaneous applications of both synthetic benchmarks and a MapReduce application on Xen-based platforms. Our experiments demonstrate the accuracy of our prediction model and indicate that Pregather results in the high disk spatial locality and a significant improvement in disk throughput and application performance.

Consolidation of Virtual Machines (VMs) on the minimum number of physical servers has been recognized as a very efficient approach to increase the efficiency of virtualized data centers and save energy, as consolidation allows unloaded servers to be switched off or used to accommodate more load. The problem is so complex that centralized and deterministic solutions are useless in large data centers with hundreds or thousands of servers. This paper presents a self-organizing approach for the consolidation of VMs on two resources, CPU and RAM. Decisions on the assignment and migration of VMs are driven by probabilistic processes and are based on local information, which makes the solution simple to implement and scalable. Experiments on a real data center show that the approach rapidly consolidates the workload, and CPU-bound and RAM-bound VMs are balanced, so that both resources are exploited efficiently. © 2013 IEEE.

This paper show covert channels using the CPU load are possible between clients connected to a multicore remote server.

Data congestion and network delay are the important factors that affect performance of cloud computing systems. Using local disk of computing nodes as a cache can sometimes get better performance than accessing data through the network. This paper presents a storage cache placement algorithm—C-Aware, which traces history access information of cache and data source, adaptively decides whether to cache data according to cache media characteristic and current access environment, and achieves good performance under different workload on storage server. We implement this algorithm in both simulated and real environments. Our simulation results using OLTP and WebSearch traces demonstrate that C-Aware achieves better adaptability to the changes of server workload. Our benchmark results in real system show that, in the scenario where the size of local cache is half of data set, C-Aware gets nearly 80% improvement compared with traditional methods when the server is not busy and still presents comparable performance when there is high workload on server side.

Professionals in various fields such as medical imaging, biology, and civil engineering require rapid access to huge amounts of pixmap image data. Multimedia interfaces further increase the need for large image databases. To fulfill these requirements, the GigaView parallel image server architecture relies on arrays of intelligent disk nodes, each disk node being composed of one processor and one disk. This contribution reviews the design of the GigaView hardware and file system, compares it to other storage servers available on the market, and evaluates fields of applications for the architecture.

International audience; In recent years, numerous applications have been deployed into mobile devices. However, until now, there have been no attempts to run simulations on handheld devices. We want investigate different architectures for running and managing simulations on handheld devices, and putting the simulation services in the Cloud. We propose a hybrid simulation and visualization approach, where a dedicated mobile application is running on the client side and the RISE simulation server is hosted in the Cloud. In particular, with our prototype, we explore the remote management of a simulation tool using a dedicated native application running on an Android Smartphone, and showing the evolution of a simulation model for a forest fire spread, mashing-up the generated graphics with online GIS services.

Data center workload prediction is important to take decisions in resources management system. Seasonal ARIMA model provide a good server workload methodology for the server workload forecasting. A large set of our experiments confirm that it has high performance, scalability and reliability and will bee integrated in our system. This paper presents a general expression in development of our forecast model in the project EnergeTic-FUI, France.

General-purpose computing domain has experienced strategy transfer from scale-up to scale-out in the past decade. In this paper, we take a step further to analyze ARM-processor based cluster against Intel X86 workstation, from both energy-efficiency and cost-efficiency perspectives. Three applications are selected and evaluated to represent diversified applications, including Web server throughput, in-memory database, and video transcoding. Through detailed measurements, we make the observations that the energy-efficiency ratio of the ARM cluster against the Intel workstation varies from 2.6-9.5 in in-memory database, to approximately 1.3 in Web server application, and 1.21 in video transcoding. We also find out that for the Intel processor that adopts dynamic voltage and frequency scaling (DVFS) techniques, the power consumption is not linear with the CPU utilization level. The maximum energy saving achievable from DVFS is 20%. Finally, by utilizing a monthly cost model of data centers, we conclude that ARM cluster based data centers are feasible, and are advantageous in computationally lightweight applications, e.g. in-memory database and network-bounded Web applications. The cost advantage of ARM cluster diminishes progressively for computation-intensive applications, i.e. dynamic Web server application and video transcoding, because the number of ARM processors needed to provide comparable performance increases.

A service provisioning system is examined, where a number of servers are used to offer different types of services to paying customers. A customer is charged for the execution of a stream of jobs; the number of jobs in the stream and the rate of their submission is specified. On the other hand, the provider promises a certain quality of service (QoS), measured by the average waiting time of the jobs in the stream. A penalty is paid if the agreed QoS requirement is not met. The objective is to maximize the total average revenue per unit time. Dynamic policies for making server allocation and stream admission decisions are introduced and evaluated. The results of several simulations are described., 8 pages, 5 figures, 16th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS 2008), pp155-162 (Best Paper Award)

Multi-step ahead response time prediction of CPU constrained computing systems is vital for admission control, overload protection and optimization of resource allocation in these systems. CPU constrained computing systems such as web servers can be modeled as single server queuing systems. These systems are stochastic and nonlinear. Thus, a well-designed nonlinear prediction scheme would be able to represent the dynamics of such a system much better than a linear scheme. A nonlinear autoregressive neural network with exogenous inputs based multi-step ahead response time predictor has been developed. The proposed estimator has many promising characteristics that make it a viable candidate for being implemented in admission control products for computing systems. It has a simple structure, is nonlinear, supports multi-step ahead prediction, and works very well under time variant and non-stationary scenarios such as single server queuing systems under time varying mean arrival rate. Performance of the proposed predictor is evaluated through simulation. Simulations show that the proposed predictor is able to predict the response times of single server queuing systems in multi-step ahead with very good precision represented by very small mean absolute and mean squared prediction errors.

International audience; Requests scheduling in Web servers is a hot research topic. Many works aim at providing optimal algorithms according to various metrics. Most of these works are based on classical scheduling metrics, considering jobs completion times, but ignoring intermediate states. We claim that this choice conduces to the design of algorithm that do not efficiently share the system resources. Indeed, Web servers have some properties that make them different than the system considered in usual scheduling theory. The classical round-robin policy, used in most production Web servers, has intrinsic qualities: it shares equally the system resources and avoids any job starvation. We introduce a novel parameterizable algorithm proposing a compromise between the benefits of the round-robin and the policies that provide the best performances. Then, we discuss the appropriate choice of the parameter depending in the requirements and the context of the Web server.

International audience; Intrusion detection systems (IDS) are one way to tackle the increasing number of attacks that exploit software vulnerabilities. However, the construction of such a security system is a delicate process involving: (i) the acquisition of the monitored program behavior and its storage in a compact way, (ii) the generation of a monitor detecting deviances in the program behavior. These problems are emphasized when dealing with complex or parallel programs. This paper presents a new approach to automatically generate a dedicated and customized IDS from C sources targeting multi-threaded programs. We use Petri Nets to benefit from a formal description able to compactly describe parallel behaviors. Obtained models can then be enhanced with extra requirements such as resources usage limits or temporal execution bounds by means of observers. We illustrate the benefits of our approach on a recent class of attacks targeting web servers.