Replica Servers Research Articles

Огляд концепції мереж доставки контенту

Article reviews the principles underlying the construction of technology content delivery networks (CDN). Optimum efficiency and reliability CDN depends on several factors. Chief among them are: the mechanisms used colocation replica to smooth traffic peaks; mechanisms for updating content, designed to automatically check the host site for changes and receive updated content for delivery to the server; mechanisms active measurement to determine the shortest path from the user to the replica server for any traffic and selection mechanisms replicas.Analyzed solutions available today. Most promising among them are those which can analyze the content delivered. New protocols and integrated solutions should be in the network and on the server to distribute, search and retrieval of content across the Internet.

Self-adaptive deployment of services in mobile environments: a study of the communication reliability on the host election algorithm

Self-adaptive service deployment and replication are one of the possible approaches to address the changing computational conditions of mobile-based environments in order to ensure certain quality attributes of the system. The host election for the dynamic placement of service replicas is one of the main steps in the process to ensure the availability, reliability and performance in mobile distributed systems. The reliability of the host election algorithm is essential for the proper functioning of these systems. However, its distributed execution may be affected by the dynamic conditions of the mobile network. This paper presents an analysis of the reliability of an election algorithm on the basis of two different transmission protocols: TCP, a reliable protocol, and UDP, a non-reliable transmission protocol. The election algorithm is part of a self-adaptive architecture designed to address the availability of the services deployed in dynamic mobile network environments.

Uniform load sharing on a hierarchical content delivery network interconnection model

Content management system (CMS) is an infrastructure for efficient distribution, organization, and delivery of digital content. It is desirable that the content must be successfully delivered regardless of the end users location or attachment network. For the end to end delivery of content, a virtual open content delivery infrastructure is formed by interconnecting several CDNs. In this paper, we focus on content delivery network interconnection. An efficient and suitable to implement hierarchical CDNI architecture, named as HCDNI, is proposed to reduce the limitations of CDNIs. Next, a content distribution and redistribution scheme is proposed so that the searching time and the round trip time for the content delivery can be minimized. Next, we find a reliable and fault tolerant scheme for web server replica placement and content caching. Finally, analysis and simulation studies show that proposed algorithm results in a significant improvement in terms of data routing, path selection, content distribution and redistribution, load balancing and network scalability.

A replicas placement approach of component services for service-based cloud application

Placement of component service replicas for service-based application (SBA) in cloud environments has become increasingly important. A SBA is usually communication topology-aware, and component service replicas possess stronger data dependency than data replicas; therefore, there are huge amounts of communication between the computer nodes that are used to place component service replicas. Because the conventional methods do not consider the communication topology of component services and the relations between computer nodes, they are not appropriate for placing component service replicas. In this paper, we propose a topological matching-based component service replicas placement method that takes into account not only the topology of SBAs but also the communication performance between different computing nodes. This method first discovers the communication topology of a SBA via multi-scale graph clustering then acquires the topology of computer nodes through spectral clustering. It then places the component service replicas by matching the above two topological structures. Comprehensive experiments are conducted by comparing the performance of our method with those of other methods based on CloudSim simulation software. The results show the effectiveness of our method for improving the performance of SBAs.

Design and Implementation of Disaster Aware Routing Protocol to Large Scale Disaster Network

Disaster resilient Network is an open issue in routing research as it occurs due to data loss and communication link disturbance in the Mobile Adhoc Network during natural disaster. The Disaster management scheme can be supported with advancement in information and communication technology (ICT) as it gains high popularity and significant research has been devoted to support loss less communication. To achieve the disaster recovery, the solution should be reliable and fast. In this paper, we propose a Disaster aware routing protocol (DRP) which can be achieved through placement of the service node repositories and replica node generation, with movable and deployable units (MDRU) to enhance the lifetime of the network. The Service node repository is modelled with different configuration of MDRU in clusters to make it as reserved nodes for disaster region. It is used to establish the node in the node failure region from the MDRU clusters to inherit invalid reservation and race condition problems. The disaster region is estimated based on the node density difference, within the bounded space. This constraint is handled by hierarchical clustering mechanism based on QOS theory by monitoring the node's transmission range, network traffic and node weight in order to reduce the node discovery and node deployment time by managing the nodes in terms of power level and bandwidth utilization. It also includes the flexible deployment protocol in order to effectively drive decision making process which capable of adapting self deployment according to different situations in the large scale disaster network. Experimental results prove that system is capable of tolerating the fault, during the natural disaster against the state art of approaches.

Joint server selection and replica placement in urban content delivery networks

In this article, we formulate the problem of optimum joint replica server deployment and content placement in urban content delivery networks as a bi-objective binary integer programming model namely JSSRPP. The proposed formulation results in an optimum design such that the miss ratio of requests, the client response time and the cost of server deployment are minimised. In practice, the popularity of files may change over the course of time, thus, a novel adaptive file replacement algorithm namely UDCR is also proposed in which files are scored according to different criteria including the time and the number of recent requests and the size of requested files. Then, in discrete points of time, the files with the lowest score are replaced with the one experiencing the highest number of misses. Results of extensive simulation study with NS-2, confirms that the hit ratio of UDCR approaches that of JSSRPP in steady state.

Virtual Servers Co-Migration for Mobile Accesses: Online versus Off-Line

In this paper, we study the problem of co-migrating a set of service replicas residing on one or more redundant virtual servers in clouds in order to satisfy a sequence of mobile batch-request demands in a cost effective way. With such a migration, we can not only reduce the service access latency for end users but also minimize the network costs for service providers. The co-migration can be achieved at the cost of bulk-data transfer and increases the overall monetary costs for the service providers. To gain the benefits of service migration while minimizing the overall costs, we propose a co-migration algorithm Migk for multiple servers, each hosting a service replicas. Migk is a randomized algorithm with a competitive cost of $O(\frac{\gamma\, \log \,n}{\min \lbrace \frac{1}{\kappa },\frac{\mu }{\lambda \,+\,\mu }\rbrace })$ to migrate $\kappa$ services in a static $n$ -node network where $\gamma$ is the maximal ratio of the migration costs between any pair of neighbor nodes in the network, and where $\lambda$ and $\mu$ represent the maximum wired transmission cost and the wireless link cost respectively. For comparison, we also study this problem in its static off-line form by proposing a parallel dynamic programming (hereafter DP) based algorithm that integrates the branch&bound strategy with sampling techniques in order to approximate the optimal DP results. We validate the advantage of the proposed algorithms via extensive simulation studies using various requests patterns and cloud network topologies. Our simulation results show that the proposed algorithms can effectively adapt to mobile access patterns to satisfy the service request sequences in a cost-effective way.

Optimistic Byzantine fault tolerance

The primary concern of traditional Byzantine fault tolerance is to ensure strong replica consistency by executing incoming requests sequentially according to a total order. Speculative execution at both clients and server replicas has been proposed as a way of reducing the end-to-end latency. In this article, we introduce optimistic Byzantine fault tolerance. Optimistic Byzantine fault tolerance aims to achieve higher throughput and lower end-to-end latency by using a weaker replica consistency model. Instead of ensuring strong safety as in traditional Byzantine fault tolerance, nonfaulty replicas are brought to a consistent state periodically and on-demand in optimistic Byzantine fault tolerance. Not all applications are suitable for optimistic Byzantine fault tolerance. We identify three types of applications, namely, realtime collaborative editing, event stream processing, and services constructed with conflict-free replicated data types, as good candidates for applying optimistic Byzantine fault tolerance. Furthermore, we provide a design guideline on how to achieve eventual consistency and how to recover from conflicts at different replicas. In optimistic Byzantine fault tolerance, a replica executes a request immediately without first establishing a total order of the message, and Byzantine agreement is used only to establish a common state synchronization point and the set of individual states needed to resolve conflicts. The recovery mechanism ensures both replica consistency and the validity of the system by identifying and removing the operations introduced by faulty clients and server replicas.

Selective partial recovery optimisation strategy for SSL connection migration

The secure socket layer SSL is a popular tool to provide data confidentiality and integrity between two communicating entities. We present a set of techniques for providing recovery strategy for SSL connection migration across the replica servers. The proposed selective partial recovery SPR optimisation strategy for SSL connection migration can be used to speedup SSL recovery session by reducing the recovery time. The strategy described is transparent to the client. The strategy can be integrated into the existing infrastructure without requiring any changes either to the TCP/IP protocol, SSL protocol, or the client. Finally, the proposed strategy is evaluated to be practical and efficient through experiment studies.

Data placement using Dewey Encoding in a hierarchical data grid

Nowadays, data grids can be seen as frameworks responding to the needs of large scale applications by affording so many resources. These are distributed on different geographically locations, but are organized to provide an integrated service. This paper studies replica placement in hierarchical data grid systems taking into account the following important issue: The hierarchical data grid can be presented as a tree structure. This structure will be labeled by a new labeling scheme (called Dewey Encoding) that lies at the core of query processing for many XML databases. This scheme avoids unnecessary replications and achieves good performance by balancing the load of replica servers. The good performance of the proposed algorithm is based on minimizing the Mean Job Execution time and Network utilization.

A novel fairness-aware parallel download scheme

Parallel download protocols that establish multiple TCP connections to distributed replica servers have the potential to reduce file download times and to achieve server-side load balancing. Unfortunately, parallel download protocols are inherently unfair towards single-flow downloads and may even starve them. This paper presents a fairness-aware parallel download protocol that allows the dynamic mitigation of throughput and fairness in distributed networks. The key novelty is a receiver-based framework that allows a dynamic adjustment of the congestion and rate control of each subconnection. The proposed Fairness-aware Control Coordinator (FCC) adjusts the priority of the subconnection between the TCP-fair rate (high priority) and the TCP-LP fair rate (low priority) according to the dynamics on the network without adding to the unfairness towards other downloads. Simulation results show the flexibility of the proposed protocol in balancing performance and fairness.

Estimation of Worst-Case Bandwidth Requirements of Video-on-Demand Systems with Replica Servers using the M/G/∞ Model

This study assesses the worst-case demand and replication models for determining the bandwidth required by the origin server when a number of video programs in a video-on-demand system are replicated on replica servers. The bandwidth required by the video- on-demand system determines the number of viewers that could watch video programs simultaneously. If a video program is viewed in its entirety, the required bandwidth is estimated based on the number of concurrent viewers as well as the length of the video program. Unlike traditional research models that use the hit ratio to assess the replication benefits, this study uses the M/G/∞ model to examine the related phenomenon and problems in this field. We develop a selection model for replication using the product of the viewing request probability and the video length as the main factor in determining the videos selected for replication. This model theoretically outperforms traditional models that consider only the viewing request probability in the benefits of reducing the worst-case demand at origin server, and the simulation results validate the theoretical results. Due to resource constraints in the replica server, discussions and verifications on the optimization of the proposed replication model were carried out and the findings were used as the basis for the development of a heuristic algorithm. The results derived using this heuristic algorithm are almost the same as the optimized results derived from LINGO.

Hash-based Byzantine fault tolerant agreement with enhanced view consistency

This paper presents an authentication-based Byzantine fault tolerant agreement protocol that is capable of detecting the tentative failures in transaction processing systems. The protocol employs a cryptographic hash function to detect crash as well as a few more types of commonly faced arbitrary behaviours manifested by the participating server replicas in the system. We show that the use of hash functions can successfully handle various types of arbitrary faults. In addition, it suggests a preemptive mechanism to proactively detect the malicious replicas. The protocol avoids frequent view changes that contributes to the liveness of the system and ensures continuity in transaction processing.

Joint Design on DCN Placement and Survivable Cloud Service Provision over All-Optical Mesh Networks

Cloud services based on data center networks (DCNs) require a transmission infrastructure with high-capacity, low-latency, low-cost and high-availability, which can be offered by survivable optical networks. DCN placement is a fundamental issue in supporting cloud services in optical networks. It concerns not only the cost of providing cloud services, but also the service availability against failures via proper service replicas. In this paper, we jointly optimize DCN placement with service routing and protection to minimize the network cost, while ensuring fast protection of all services against any single link failure or service failure at a particular DCN. An ILP (Integer Linear Program) is first formulated to achieve optimal joint design. It integrates p-cycle (preconfigured protection cycle) for fast protection against a single link failure, and DCN replicas and fast service rerouting against a service failure. To make the design more scalable, a two-step heuristic is then proposed for large-size network scenarios. The first step separately solves the DCN placement and service routing problem in the failure-free scenario, and the second step takes fast service protection into account. The proposed design is validated by extensive numerical experiments.

A divide-and-conquer approach for content replication in WMNs

Wireless Mesh Networks (WMNs) extend Internet access in areas where the wired infrastructure is not available. A problem that arises is the congestion around gateways, delayed access latency and low throughput. Therefore, object replication and placement is essential for multi-hop wireless networks. Many replication schemes are proposed for the Internet, but they are designed for CDNs that have both high bandwidth and high server capacity, which makes them unsuitable for the wireless environment. Object replication has received comparatively less attention from the research community when it comes to WMNs. In this paper, we propose an object replication and placement scheme for WMNs. In our scheme, each mesh router acts as a replica server in a peer-to-peer fashion. The scheme exploits graph partitioning to build a hierarchy from fine-grained to coarse-grained partitions. The challenge is to replicate content as close as possible to the requesting clients and thus reduce the access latency per object, while minimizing the number of replicas. Using simulation tests, we demonstrate that our scheme is scalable, performing well with respect to the number of replica servers and the number of objects. The simulation results show that our proposed scheme has better performance compared to other replication schemes.

ILP Modeling of Many-to-Many Replicated Multimedia Communication

On-line communication services were evolving from a simple text-based chats towards sophisticated videopresence appliances. The bandwidth consumption of those services is constantly growing due to the technology development and high user and business needs. That fact leads us to implement optimization mechanisms into the multimedia communication scenarios. In this paper, the authors concentrate on many-to-many (m2m) communication, that is mainly driven by the growing popularity of on-line conferences and telepresence applications. An overlay model where m2m flows are optimally established on top of a given set of network routes is formulated and a joint model where the network routes and the m2m flows are jointly optimized. In the models, the traffic traverses through replica servers, that are responsible for stream aggregation and compression. Models for both predefined replica locations and optimized server settlement are presented. Each model is being followed by a comprehensive description and is based on real teleconference systems.

A heuristic approach to working and spare capacity optimization for survivable anycast streaming protected by p-cycles

Most of previous research in the field of network survivability has been focused on unicast transmissions. However, growing popularity of various concepts following the idea of content-oriented networks has triggered the need to develop new approaches to protect networks with other than unicast flows including multicast and anycast flows. In this paper, we consider the latter case and address the problem of working and spare capacity allocation in networks with anycast streaming, i.e., it is assumed that a set of replica servers is deployed in the network to serve streaming requests. To protect the network we propose to apply p-cycles--a relatively novel survivability approach combining capacity effectiveness of mesh restoration and ring-like restoration speed. To benefit from special properties of anycast flows, we use augmented version of p-cycles called Anycast-Protecting p-Cycles (APpC). We formulate the optimization problem as an ILP (Integer Linear Programming) model further applied to obtain optimal results using the GuRoBi solver. Due to high complexity of this optimization problem, we propose an effective heuristic algorithm based on the Simulated Annealing approach. Several versions of the algorithm are developed and examined--the best method yields on average results only 4.14 % worse than optimal ones. A wide range of experiments is conducted to verify performance of the proposed approach as a function of various network parameters including: p-cycle generator, p-cycle length, number of replica servers, number of clients. Moreover, we evaluate the Anycast-Protecting p-Cycles approach against classical p-cycle.

Selfish Less Replica Allocation in MANET

of the protocols and algorithms used in MANET, are assuming that all mobile nodes cooperate fully with the functionalities of the network. But some nodes are cooperate partially or not at all cooperating. Network performance and data accessibility, accessing time, query delay are affected by these selfish nodes. The discussion of this paper is about, representing a Replica server which solves the selfishness. The Replica server monitors and maintains the status of all mobile nodes in the network. If it finds any selfish node in the shortest path between source and destination, the replica server will analyze the reason for selfishness and it finds solution. The conducted simulations demonstrate the proposed approach based on proxy method which outperforms in terms of network parameters such as accessing data, time and cost and also it improves the network performance of MANET. Keywordsdata accessibility, accessing time, query delay, selfish nodes, Replica server, proxy method.

Request Redirection Policy Based on Service Utility in CDN

To quantify the service utility of CDN, some metrics is developed to measure its content-serving utility. The quantitative measure can provide an indication of the perceived benefits of CDN system and express the service effectiveness of each replica server. The utility measure is used to devise a request-redirection policy and conduct extensive simulations to demonstrate that the approach can improve content delivery performance in CDN.

Optimal configuration of fault-tolerance parameters for distributed replicated server access

Server replication is a common fault-tolerance strategy to improve transaction dependability for services in communications networks. In distributed architectures, fault-diagnosis and recovery are implemented via the interaction of the server replicas with the clients and other entities such as enhanced name servers. Such architectures provide an increased number of redundancy configuration choices. The influence of a (wide area) network connection can be quite significant and induce trade-offs between dependability and user-perceived performance. This paper develops a quantitative stochastic model using stochastic activity networks (SAN) for the evaluation of performance and dependability metrics of a generic transaction-based service implemented on a distributed replication architecture. The composite SAN model can be easily adapted to a wide range of client-server applications deployed in replicated server architectures. In order to obtain insight into the system behaviour, a set of relevant environment parameters and controllable fault-tolerance parameters are chosen and the dependability/performance trade-off is evaluated.