Service Replication Research Articles

On joint routing and server selection for MD video streaming in ad hoc networks

For media streaming in ad hoc networks, service replication has been demonstrated to be a quite effective countermeasure to streaming interruptions caused by fragile paths and dynamic topology. In this paper, we study the problem of joint routing and server selection for double description (DD) video streaming in ad hoc networks. We formulate the task as a combinatorial optimization problem and present tight lower and upper bounds for the achievable distortion. The upper bound provides a feasible solution to the formulated problem. Our extensive numerical results show that the bounds are very close to each other for all the cases studied, indicating the near-global optimality of the derived upper bounding solution. Moreover, we observe significant gains in video quality achieved by the proposed approach over existing server selection schemes. This justifies the importance of jointly considering routing and server selection for optimal MD video streaming

Design and implementation of a secure wide-area object middleware

Wide-area service replication is becoming increasingly common, with the emergence of new operational models such as content delivery networks and computational grids. This paper describes the security architecture for Globe, an object-based middleware specifically designed to support dynamic replication of services over wide-area networks. Replication introduces a series of new security issues, including the need to restrict replica privileges with respect to method execution, and protection of distributed objects against malicious hosts running instances of their code. Our modular security design addresses these new threats, as well as a broad range of traditional ones, and is validated through a series of performance measurements. Additional contributions include a novel authentication mechanism specifically designed for wide-area deployment, which combines some of the best features of public key authentication protocols (reliance on an offline trusted third party in particular) with the computational efficiency characteristic to symmetric key schemes.

Optimal Server Replication Schemes to Reduce Location Management Cost in Cellular Network

The default server strategy is commonly used to manage the location and state of mobile hosts in cellular networks. With this strategy, connections can be established after the client obtains the location information of the mobile host by querying the default server. Unfortunately, the communication cost increases if the query requests are frequent and the distance between the default server and the client is long. Still more, no connection to a mobile host can be established when the default server of the destination mobile host fails. These problems can be solved by replicating the default servers and by letting the nearest replicated default server process the query request which is sent from a client [9]. It is important to allocate replicated default servers efficiently in networks and determine the number of replicated default servers. In this paper, we suggest and evaluate a default server replication strategy to reduce communication costs and to improve service availabilities. We consider optimal replication degree as well as location for replicating the default servers in n-grid and tree networks.

Distributed Wisdom: Designing a Replication Service for Large Peer-to-Peer Data Grids

We discuss the peer-to-peer middleware architecture for developing grid applications. We present the key issues in building services for large-scale distributed systems. Our approach of developing a replication service using a scalable, reconfigurable shared-object space realized over two-layer P2P architecture appears unique. However, we need to benchmark the platform to test the performance. Also, we must look at key issues in designing the replication service itself, including data redundancy and efficient data placement strategies to minimize computation time and bandwidth.

A replication framework for program‐to‐program interaction across unreliable networks and its implementation in a servlet container

AbstractWe propose a service replication framework for unreliable networks. The service exhibits the same consistency guarantees about the order of execution of operation requests as its non‐replicated implementation. Such guarantees are preserved in spite of server replica failure or network failure (either between server replicas or between a client and a server replica), and irrespective of when the failure occurs. Moreover, the service guarantees that in the case when a client sends an ‘update’ request multiple times, there is no risk that the request be executed multiple times. No hypotheses about the timing retransmission policy of clients are made, e.g. the very same request might even arrive at different server replicas simultaneously. All of these features make the proposed framework particularly suitable for interaction between remote programs, a scenario that is gaining increasing importance. We discuss a prototype implementation of our replication framework based on Tomcat, a very popular Java‐based Web server. The prototype comes into two flavors: replication of HTTP client session data and replication of a counter accessed as a Web service. Copyright © 2005 John Wiley & Sons, Ltd.

A pull‐based service replication protocol in mobile ad hoc networks

AbstractThe mobility of nodes in ad hoc networks can lead to frequent network partitioning. This partitioning disconnects many nodes from a centralised server. Being aware of a future partitioning can help to provide a continuous service availability for all mobile nodes. In this paper, we propose a service replication protocol consisting of three distributed algorithms. This protocol (i) predicts the occurrence of network partitioning, which allows to create service replicas in the future separate partitions and (ii) constructs a dynamic service replicas deployment scheme. Using linear‐time temporal logic, we show that each detection of network partitioning can be predicted before its occurrence. Our simulation results show that the proposed protocol guarantees service availability to all mobile nodes without incurring high cost. Copyright © 2005 AEIT.

Enabling Grid Access to Mass Storage

The purpose of the EDG Storage Element (SE) is to provide a uniform Grid interface to mass storage systems. We needed an extensible and flexible architecture, and chose one based on message processing. Not only has this architecture provided a solid framework for meeting changing requirements, it also provides scope for building an extensible and scalable storage management system. While distributed SEs in a Grid are managed by higher level replication services, our request processing model opens possibilities for distributing a single SE across several nodes, allowing further opportunities for optimisation. In this paper, we discuss applications of distributed, transactional processing in mass storage, and its implications for optimisations based on access patterns.

Content Delivery Policies in Replicated Web Services: Client-Side vs. Server-Side

Replication of Web Services has an important role among techniques that have been developed in order to meet the demand for faster and more efficient access to the Internet. Replication can be addressed both by a cluster of servers, and by servers geographically distributed in the Internet. In this paper, we focus on geographical replication. Two approaches are commonly used for geographical replication: server-side and client-side. In the client-side approach, the client has a significant role in the policy used to exploit Web Service Replication. In the server-side approach, the client transparently exploits a Replicated Web Service. Both server-side and client-side approaches provide various strategies that can be adopted. An analysis of these strategies, and an overall classification, is presented here. Finally, client-side and server-side approaches are compared, identifying their pros and cons in order to propose the features of an eventual complete approach.

Roam: A Scalable Replication System for Mobility

Nomadic users require replication to store copies of critical data on their mobile machines while disconnected or poorly connected. Existing replication services do not provide all classes of mobile users with the capabilities they require, which include: the ability for direct synchronization between any two replicas, support for large numbers of replicas, and detailed control over what files reside on their local (mobile) replica. Mobile users must adapt their behavior to match the level of service provided by today's replication systems, thereby hindering mobility and costing additional time, money, and systems management. Roam is a replication system designed to satisfy the requirements of the mobile user. Roam is based on the Ward Model, a replication architecture for mobile environments. Using the Ward Model and new distributed algorithms, Roam provides a scalable replication solution for the mobile user. We describe the motivation, design, and implementation of Roam and report its performance.

Design and analysis of a replicated elusive server scheme for mitigating denial of service attacks

The paper proposes a scheme, referred to as proactive server roaming, to mitigate the effects of denial of service (DoS) attacks. The scheme is based on the concept of “replicated elusive service”, which through server roaming, causes the service to physically migrate from one physical location to another. Furthermore, the proactiveness of the scheme makes it difficult for attackers to guess when or where servers roam. The combined effect of elusive service replication and proactive roaming makes the scheme resilient to DoS attacks, thereby ensuring a high-level of quality of service. The paper describes the basic components of the scheme and discusses a simulation study to assess the performance of the scheme for different types of DoS attacks. The details of the NS2-based design and implementation of the server roaming strategy to mitigate the DoS attacks are provided, along with a thorough discussion and analysis of the simulation results.

Nonstop: continuous multimedia streaming in wireless ad hoc networks with node mobility

Guaranteeing continuous streaming of multimedia data from service providers to the users is a challenging task in wireless ad hoc networks, particularly when node mobility is considered. The topological dynamics introduced by node mobility are further exacerbated by the natural grouping behavior of mobile users, which leads to frequent network partitioning. Network partitioning poses significant challenges to the provisioning of continuous multimedia streaming services in wireless ad hoc networks, since the partitioning disconnects many mobile users from the centralized streaming service. In this paper, we propose NonStop, a collection of novel middleware-based run-time algorithms that ensures the continuous availability of such multimedia streaming services, while minimizing the overhead involved. The network-wide continuous streaming coverage is achieved by partition prediction and service replication on the streaming sources and assisted by distributed selection of streaming sources on regular mobile nodes and users. The proposed algorithms are validated by extensive results from performance evaluations.

서비스 거부 공격에 대비한 자원 재할당 및 서버 중복 방안

컴퓨팅 노드가 가지고 있는 자원을 고갈시켜서 본래 의도한 서비스 제공을 방해하려는 서비스 거부 공격에 대응하기 위해서는 어떠한 경우에도 반드시 유지되어야 하는 필수 서비스를 위한 중요 자원을 식별하고 위급 상황에 적절히 시스템을 적응시키고 재구성하는 방안이 필요하다. 본 논문에서는 이를 위하여 두 단계의 대응 방안을 제시한다. 첫 번째 단계로 한 컴퓨팅 노드 내에서 선택된 필수 서비스에 대해 자원을 동적으로 할당하여 공격이 성공한 후에도 필수 서비스가 유지될 수 있도록 한다. 이 조치에도 불구하고 노드 내에서의 충분한 자원 확보가 불가능해지면 두 번째 단계로 미리 준비된 다른 컴퓨팅 노드에서 사용자에게 투명하게 필수 서비스가 제공될 수 있도록 중복성을 적용하는 방안을 제시한다. 테스트베드를 구축하여 실험을 실시한 결과 기법의 타당성을 입증할 수 있었다. 기존의 다른 방법과 성능 및 비용을 분석하여 비교하였다. In order to cope with DoS (Denial of Service) attacks disturbing delivery of intended services by exhausting resources of computing nodes, we need a solution to recognize important resources for the essential services which have to be maintained under any circumstances and to adapt the system to the urgent situation and reconfigure itself properly. In this paper, we present a two-phase scheme to handle the problem. In the first phase, by means of dynamic resource reallocation within a computing node, we try to make the selected essential services survive even after the occurrence of an attack. For the second phase when it becomes impossible to continue the service in spite of the actions taken in the first phase, we apply server replication in order to continue the transparent provision of the essential services with the end users by utilizing redundant computing nodes previously arranged. Experimental result obtained on a testbed reveals the validity of the proposed scheme. A comparison with other proposed schemes has been conducted by analyzing the performance and the cost.

A Dynamic Parallel Data-Computing Environment for Cross-Sensor Satellite Data Merger and Scientific Analysis

Abstract The Data Processing and Error Analysis System (DPEAS) is a dynamic parallel data processing system for cross-sensor satellite data merger and analysis. Using a peer-to-peer methodology, DPEAS is able to distribute and simplify the near-real-time merger of multiple datasets from several sensors and satellite platforms. The organization of the system and its practical benefits for research and operational use are discussed. Using several simplifying constructs, the technical requirements placed upon the end user are minimized, while maintaining high scalability and reliability of the system. Thus, the system encourages the flow of research into operations by making the system environment mobile, rather than placing the transferability burden on the research development effort. Several reliability enhancements, such as cluster failover and automatic file replication services, are incorporated into the system. The technical aspects of the various feature implementations are discussed in detail. Futur...

The evolution of Coda

Failure-resilient, scalable, and secure read-write access to shared information by mobile and static users over wireless and wired networks is a fundamental computing challenge. In this article, we describe how the Coda file system has evolved to meet this challenge through the development of mechanisms for server replication, disconnected operation, adaptive use of weak connectivity, isolation-only transactions, translucent caching, and opportunistic exploitation of hardware surrogates. For each mechanism, the article explains how usage experience with it led to the insights for another mechanism. It also shows how Coda has been influenced by the work of other researchers and by industry. The article closes with a discussion of the technical and nontechnical lessons that can be learned from the evolution of the system.

Secure dynamic anycasting for ‘ best’ server selection using active networks

The networking community has tackled the resource-finding problem using several methods. The knowledge of the name or property of the resource enables one to find it over the network. Many techniques were proposed and investigated for a single instance of the resource. The Internet has experienced dramatic growth in the use and provision of services such as ftp, gopher, archie and World-Wide-Web. The heavy demands being placed on servers inspire replication (mirroring) of servers. This replication results in client intending to contact the ‘best’ server among many content-equivalent servers. The solutions that were used for the ‘best’ server selection include multicast and broadcast communication to send request to all servers and choose the best one from all the replies. These solutions require the client to be powerful enough to handle all the replies, which may be overwhelming leading to client's machine getting hung. The other solution uses the name servers to provide a different unicast address for one member of a group of servers at different locations. The inherent disadvantage in this method is that the user is unable to choose the best server. The idea of application layer anycasting allows the user to select the best server according to the user's selection criteria. The main disadvantage of this scheme is that the client that does the selection may not be powerful to handle responses from all the content-equivalent servers. In this paper the idea of application layer anycasting has been extended by allowing the active routers to locate the best server. Active networks, unlike the traditional networks are not just passive carrier of bits but instead provides the capability for the user to inject customized programs into the networks that may modify, store or redirect the user data flowing through the network. Anycasting is done in the application level as it provides better end-to-end control, and there is no support in the network level. The choice of ‘best’ server is done based on the first response from the servers. The active routers do the filtering of responses from laggards and the client gets the response only from the best server, thus the client machine is not overwhelmed by responses. The client deals with the vital issue of security with respect to Active networks by the use of various encryption schemes. Since the ‘best’ server chosen is not always the best forever, a TTL value is associated with each of the best server found, and the best one is reselected after its expiry. The performance of the proposed scheme is compared with the networks without active networks and is found to provide better response time for requests. Further, the proposed scheme avoids the overloading of a server, jockeying, and reduces the overhead of the client in selecting the best server. The overhead on the routers in active networks is insignificant compared to the advantages accrued due to it.

Replication requirements in mobile environments

Replication is extremely important in mobile environments because nomadic users require local copies of important data. However, today's replication systems are not "mobile-ready". Instead of improving the mobile user's environment, the replication system actually hinders mobility and complicates mobile operation. Designed for stationary environments, the replication services do not and cannot provide mobile users with the capabilities they require. Replication in mobile environments requires fundamentally different solutions than those previously proposed, because nomadicity presents a fundamentally new and different computing paradigm. Here we outline the requirements that mobility places on the replication service, and briefly describe ROAM, a system designed to meet those requirements.

DTV coverage and service prediction, measurement and performance indices

It is now clear that methods used to predict the replication of the NTSC service in the United States were idealized and, for the most part, not validated. The need to review the methods and procedures now in use is highlighted by the various field tests conducted in the U.S. and elsewhere. This paper outlines potential improvements in the modeling, methods and procedures now in use for prediction and measurement of the DTV service.

Replicated servers allocation for multiple information sources in a distributed environment

Recently, we have witnessed a phenomenal growth in the Internet/Intranet coupled with rapid deployment of new services. Information dissemination over the network has become one of the most important activities in our daily life. The existing systems, however, often suffer from notorious long delay experienced by clients, especially during peak hours. This is caused by the combination of a variety of factors including inadequate link bandwidth, server overload and network congestion. Server replication has been shown to be one of the most effective mechanisms to cope with this problem. The basic idea is to replicate the information across a network so that the clients’ requests can be spread out. The major issue is in which locations inside the network this replication takes place at, i.e. where to place the replicated servers. In this paper we investigate the server replication in a controlled distributed environment. The salient feature of this environment is that the decision of where to replicate information can be determined by a single authority, the Intranet being the typical example. We consider the problem of placing multiple replicated servers within a network, given there exist multiple target web servers as information providers. We formulate this as an optimization problem by taking into consideration the characteristics of the network topology. We first show that this is an NP-complete problem, and then present a number of heuristics-based algorithms for server replications. Finally, in order to investigate various tradeoffs in terms of cost and algorithm complexity, we carry out comparison studies among different heuristic algorithms.

Performance analysis of a scaleable design for replicating file collections in wide-area networks

Mirroring file collections in the global Internet is widely practiced with a recent study estimating the number of WWW hosts with mirrored content at 10% of all WWW hosts. Conventional mirroring tools, however, are not well-suited for the large-scale multiple-site replication services envisioned by projects such as the Internet2 Distributed Storage Infrastructure (I2-DSI) project. This paper presents a scaleable design for the automated synchronization of large collections of files replicated across multiple hosts, as in I2-DSI, and outlines of how the design has been realized using rsync+, a modification to the popular open-source mirroring tool, rsync. A performance study based on an instrumented mirror usingrsync + empirically characterizes server-side processing costs under realistic, large-scale workloads, and supplementary measurements of network throughput across Internet2 links illustrate the achievable network performance in a high-speed wide-area network. These experimental results confirm the validity of scalability arguments for the design, uncover key system parameters for rsync+ that must be tuned for efficient operation, and indicate the limitations of TCP-only transport solutions as the number of mirror sites grows.

Application-layer anycasting: a server selection architecture and use in a replicated Web service

Server replication improves the ability of a service to handle a large number of clients. One of the important factors in the efficient utilization of replicated servers is the ability to direct client requests to the "best" server, according to some optimality criteria. In the anycasting communication paradigm, a sender communicates with a receiver chosen from an anycast group of equivalent receivers. As such, anycasting is well suited to the problem of directing clients to replicated servers. This paper examines the definition and support of the anycasting paradigm at the application-layer, providing a service that uses an anycast resolver to map an anycast domain name and a selection criteria into an IP address. By realizing anycasting in the application-layer, we achieve flexibility in the optimization criteria and ease the deployment of the service. As a case study, we examine the performance of our system for a key service: replicated Web servers. To this end, we develop an approach for estimating the response time that a client will experience when accessing given servers. Such information is maintained in the anycast resolver that clients query to obtain the identity of the server with the best estimated response time. Our performance collection technique combines server push with resolver probes to estimate the expected response time without undue overhead. Our experiments show that selecting a server using our architecture and estimation technique can improve the client response time by a factor of two over nearest server selection and by a factor of four over random server selection.