Replica Selection Research Articles

NeIL: Intelligent Replica Selection for Distributed Applications

NeIL: Intelligent Replica Selection for Distributed Applications

A novel cost-based replica server placement for optimal service quality in cloud-based content delivery network

<span lang="EN-US">Replica server placement is one of the crucial concerns for a given geographic diversity associated with placement problems in content delivery network (CDN). After reviewing the existing literatures, it is noted that studies are more for solving placement problem in conventional CDN and not much over cloud-based CDN architectures, which some few studies are reported towards replica selection are much in its nascent stages of development. Moreover, such models are not benchmarked or practically assessed to prove its effectiveness. Hence, the proposed study introduces a novel design of computational framework associated with cloud-based CDN which can facilitate cost-effective replica server management for enhanced service delivery. Implemented using analytical research methodology, the simulated study outcome shows that proposed scheme offers reduced cost, reduced resource dependencies, reduced latency, and faster processing time in contrast to existing models of replica server placement.</span>

AMS: Adaptive Multiget Scheduling Algorithm for Distributed Key-Value Stores

Distributed key-value stores provide the Multiget API, where many key-value operations are batched together, to meet the parallel requirement of applications. Correspondingly, reducing the latency of Multigets is crucial for the responsiveness of the distributed key-value stores. The latency of a Multiget depends on both which replica server its key-value operations are scheduled to, i.e., the replica selection for each key-value operation, and when these key-value operations are served, i.e., the scheduling of the service sequence at different replica servers. Existing solutions solely focus on either one of them and accordingly lead to suboptimal latency for Multigets. To address these issues, this paper proposes an Adaptive Multiget Scheduling (AMS) algorithm in this paper, and specifically, our AMS re-architectures the framework to remove the conflict between replica selection and service sequence scheduling in the existing solution Rein. Based on the new framework, a sophisticated replica selection method is designed. Furthermore, AMS guides both replica selection and service sequence scheduling by the piggybacked information of replica servers, being adaptive to the heterogeneous time-varying server performance. Consequently, AMS can respectively reduce the median, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$95^{th}$</tex-math></inline-formula> , and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$99^{th}$</tex-math></inline-formula> percentile latencies of Multigets by a factor of 4, 3.1, and 1.86 compared to the default FIFO algorithm and significantly outperforms Rein.

A Prediction Based Replica Selection Strategy for Reducing Tail Latency in Geo-Distributed Systems

The deployment of modern applications in geo-distributed systems results in performance fluctuation which is a consequence of long-tail latency. To deliver high-quality services these applications always strive to adapt to the changing situation and an appropriate replica selection strategy is one efficient way to achieve this. Several replica selection strategies have already been developed but none of them are efficient enough to reduce tail latency and adapt to the dynamic environment of the geo-distributed systems. In this paper, we present the design and implementation of a prediction based replica selection strategy for reducing tail latency in geo-distributed systems. We have meticulously designed the proposed strategy to adapt to the dynamic behavior of the distributed system. For evaluating its effectiveness in reducing tail latency and adapting to the dynamic behavior of the geo-distributed systems we perform some extensive experiments in a 15 nodes Cassandra cluster that is deployed on Amazon EC2 over 5 geographical regions. For generating test datasets and workloads we use industry-standard Yahoo Cloud Serving Benchmark (YCSB). Our experimental results show that the proposed strategy not only reduces tail latency but also increases the overall throughput of the systems.

Accelerated Information Dissemination for Replica Selection in Distributed Key-Value Store Systems

In distributed key-value stores, multiple replica servers are always available for each key-value access operation when the eventual consistency model is employed. Accordingly, the completion times of the key-value access operations generated by an end-user request at different servers may be of great difference, especially when the replica servers are heterogeneous and have time-varying performance. Accordingly, the replica selection algorithm is crucial to cut the response time of end-user requests. The main challenge of making replica selection for each light-weighted key-value access operation is to timely know the status of replica serves. Recently, the adaptive replica selection algorithm C3 suggests guiding the replica selection with the piggybacked information of replica server in the returned “value”. Although C3 has good performance, the poor timeliness of feedback information makes a large performance gap between C3 and the ideal replica selection algorithm. To narrow this gap, the Accelerated Information Dissemination (AID) mechanism is proposed in this paper. Specifically, AID removes the bottleneck of information dissemination at the “slow” servers by letting both “client” and “replica server” store the records about the status of replica servers and both “key” and “value” piggyback multiple records. AID is implemented in Cassandra and evaluated by experiments and large scale simulations. The results show AID can significantly improve the timeliness of feedback information, especially when the number of nodes is large. Accordingly, AID helps C3 to greatly reduce the latency.

Artemis: A Latency-Oriented Naming and Routing System

Today, Internet service deployment is typically implemented with server replication at multiple locations. Domain name system (DNS), which translates human-readable domain names into network-routable IP addresses, is typically used for distributing users to different server replicas. However, DNS relies on several network-based queries and the queries delay the connection setup process between the client and the server replica. In this article, we propose Artemis, a practical low-latency naming and routing system that supports optimal server (replica) selection based on user-defined policies and provides lower query latencies than DNS. Artemis uses a DNS-like domain name-IP mapping for replica selection and achieves low query latency by combining the name resolution process with the transport layer handshake process. In Artemis, all server replicas at different locations share the same anycast IP address, called Service Address. Clients use the Service Address to establish a transport layer connection with the server. The client's initial handshake packet is routed over an overlay network to reach the optimal server. Then the server migrates the transport layer connection to its original unicast IP address after finishing the handshake process. After that, service discovery is completed, and the client communicates with the server directly via IP addresses. To validate the effectiveness of Artemis, we evaluate its performance via both real trace-driven simulation and real-world deployment. The result shows that Artemis can handle a large number of connections and reduce the connection setup latency compared with state-of-the-art solutions. More specifically, our deployment across 11 Google data centers shows that Artemis reduces the connection setup latency by 39.4% compared with DNS.

Intelligent Replica Selection in Edge and IoT Environments Using Artificial Neural Networks

Cloud, edge and Internet of Things (IoT) technologies have emerged to overcome the challenges involved in sharing computational resources and information services. Within generic cloud systems, two models have been identified as having widespread applicability: computation clouds and data clouds. A data cloud is cloud computing that aims to manage, unify and operate multiple data workloads. Many current applications generate datasets consisting of petabytes (PB) of information. Managing large datasets is a complex issuel; in particular, datasets associated with many applications can be distributed widely in geographical terms, particularly in IoT systems. Edge and IoT systems are facing new challenges with increased complexity, making scalability an important issue that will affect the performance of the system. Data replication services are widely accepted techniques to improve availability and fault tolerance, and to improve the data access time. Current replication services, however, often exhibit an increase in response time, reflecting the problems associated with the ever-increasing size of databases. This paper proposes a prediction model to predict replica locations using the files’ access profile, which feeds the neural networks with the access and location behavior (file profile) to minimize the overhead of transferring large volumes of data, which slows down the system and requires careful management. This new model has shown high accuracy and low overheads. The result shows a significant improvement in total task execution time using the proposed model for locating files by 16.34% and 30.45%; in addition, the results show bandwidth improvement by 24.7% and 49.4% compared to the user profile prediction model and replica service model without prediction, respectively. Consequently, the proposed algorithm can improve data access speed, reduce data access latency and decrease bandwidth consumption.

Replica selection and placement techniques on the IoT and edge computing: a deep study

Internet of Things (IoT) has lately been presented as a new technological transformation in which things are connected via the Internet. Several sensors and devices create data and send vital signals constantly over sophisticated networks that allow machine-to-machine interactions and monitor and manage key smart-world infrastructures. Since huge amounts of data are generated, reducing the data access costs is a critical issue. Edge computing has been developed as a novel paradigm for solving IoT demands to reduce the rise in resource congestion. One of the most significant data management challenges in the IoT is selecting suitable replication things that minimize reaction time and cost. Therefore, our goal is to examine replica selection and placement techniques in IoT and edge computing. The findings revealed that the edge computing environment might significantly enhance system performance regarding access response time, prediction accuracy, effective network, and increased data availability. Furthermore, the results illustrate that data provenance is necessary to raise the accuracy of the data by. Also, the results showed that the most important challenge in data replication and placement techniques in IoT and edge computing was the availability of data and access response time.

A perceptron-based replication scheme for managing the shared last level cache

The shared last level cache (SLLC), which provides large effective cache capacity, is widely adopted in modern chip multiprocessors (CMPs). But, long on-chip access latency in the SLLC is a key problem that hurts system performance. Replication is an effective way to relieve this problem through storing a replica of L1 victims in the near local LLC slice. However, previous replication schemes either blindly create replicas based on no feature of cache blocks or select replicas based on a single feature (such as data type, access count, etc.), which will affect the replication accuracy and limit the system performance improvements. In this paper, according to the successful application of machine learning (ML) in the field of computer architecture optimization in recent years, we develop a novel perceptron-based replication scheme (PBR) for effectively managing the SLLC in CMPs. Unlike existing single-feature-based schemes, this scheme effectively combines four features related to the reuse behavior of L1 victims, which are address (Addr), program counter (PC), data type (DT), and access count (AC), through perceptron to facilitate the accuracy of replica selection. Experimental results show that compared with the two previously proposed single-feature-based replication schemes: ASR and LADR, PBR decreases the execution time by 6.59% and 18.27%, and reduces the network traffic by 10.35% and 13.18% respectively with negligible energy consumption, hardware and area overhead.

An Enhanced Replica Selection Approach Based on Distance Constraint in ICN

Fifth generation (5G) networks have a high requirement for low latency of data delivery. Information-centric networking (ICN) adopts the paradigm of separation of the identifier and locator. It is efficient in content distribution by supporting in-network caching and has the potential to satisfy the low latency requirement in 5G. Replica selection is a key problem to retrieving content in ICN. Prior research usually utilizes the nearest replica. However, using the nearest replica cannot guarantee the smallest content download delay. To exploit in-network caching better, we propose an enhanced replica selection approach, called ERS. ERS first uses a distance-constrained-based name resolution system to discover the nearby replicas. Then, the most appropriate replica is chosen according to a local state table that maintains the state of replica nodes within a limited domain. In addition to network distance and replica node load, ERS innovatively introduces the path congestion degree between requester and replica nodes to assist replica selection. With extensive simulations, the proposed approach shows better performance than the state-of-the-art methods in terms of average content download delay. Finally, the overhead of the proposed method is analyzed.

A Novel Intelligent Approach for Dynamic Data Replication in Cloud Environment

In recent years, cloud computing research, specifically data replication techniques and their applications, has been growing. If the replicas number is raised and put in multiple positions, it will be expensive to maintain the data usability, performance and stability of the application systems. In this paper, two bio- inspired algorithms were proposed to improve both selection and placement of data replicas in the cloud environment. The suggested algorithms for dynamic data replication are multi-objective particle swarm optimization (MO-PSO) and ant colony optimization (MO-ACO). The first suggested algorithm, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., MO-PSO, is employed to obtain the best selected data replica depend on the most frequent one. However, the second suggested algorithm, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., MO-ACO, is employed to obtain the best data replica placement depend on the shortest distance, and the replicas availability. A simulation of the suggested strategy was carried out using CloudSim. Each data center (DC) includes hosts with set of virtual machines (VMs). The data replication order is determined at random from a thousand cloudlets. All replication files are randomly distributed in the proposed architecture. The performance of suggested techniques was evaluated against several approaches including: Adaptive Replica Dynamic Strategy (ARDS), Enhance Fast Spread (EFS), Genetic Algorithm (GA), Replica Selection and Placement (RSP), Popular File Replication First (PFRF), and Dynamic Cost-aware Re-replication and Re-balancing Strategy (DCR2S). The simulation results prove that MOPSO gives improved data replication compared against other algorithms. Additionally, MOACO realizes higher data availability, lower cost, and less bandwidth consumption compared with other algorithms.

Minimizing data access latency in data grids by neighborhood‐based data replication and job scheduling

SummaryIn Data Grid systems, quick data access is a challenging issue due to the high latency. The failure of requests is one of the most common matters in these systems that has an impact on performance and access delay. Job scheduling and data replication are two main techniques in reducing access latency. In this paper, we propose two new neighborhood‐based job scheduling strategies and a novel neighborhood‐based dynamic data replication algorithm (NDDR). The proposed algorithms reduce the access latency by considering a variety of practical parameters for decision making and the access delay by considering the failure probability of a node in job scheduling, replica selection, and replica placement. The proposed neighborhood concept in job scheduling includes all the nodes with low data transmission costs. Therefore, we can select the best computational node and reduce the search time by running a hierarchical and parallel search. NDDR reduces the access latency through selecting the best replica by performing a hierarchical search established based on the access time, storage queue workload, storage speed, and failure probability. NDDR improves the load balancing and data locality by selecting the best replication place considering the workload, temporal locality, geographical locality, and spatial locality. We evaluate our proposed algorithms by using Optorsim Simulator in two scenarios. The simulations confirm that the proposed algorithms improve the results compared with similar existing algorithms by 11%, 15%, 12%, and 10% in terms of mean job time, replication frequency, mean data access latency, and effective network usage, respectively.

A metaheuristic‐based method for replica selection in the Internet of Things

SummaryWith the advent of the Internet of Things (IoT), the count of gadgets connected to the Internet has been increased. IoT, as a modern paradigm, has been used to describe the future in which physical things like RFID tags, sensors, actuators, and cellphones can intermingle for achieving shared purposes. Also, we can employ cloud computing for storing the things' information in the IoT. However, this information has been replicated through the network for increasing availability. In this paper, due to the NP‐hard nature of the replica selection problem, an improved version of ant colony optimization (IACO) has been applied. The impact of pheromone on the chosen path is converted by ants to invert the underlying logic of ACO. Due to the existence of different IoT centers, the IACO has been employed for selecting the replicated data in the IoT where the load balancing among IoT centers has been considered. In this method, an ant chooses the ideal point for its movement; then others may not pass the track that the preceding ants have been passed. The obtained outcomes have shown that the method has outperformed the ACO, HQFR, and RTRM approaches regarding the waiting time and load balancing.

Multiple users replica selection in data grids for fair user satisfaction: A hybrid approach

Replica selection in data grids aims to select the best replica location based on the quality-of-service parameters preferred by the user. This choice is important because of the limited number of available data resources in comparison with the large number of users. Typically, user requests are fulfilled in a first-in, first-out manner. This may satisfy the users at the beginning of the queue more than those at the end. Better results can be achieved by considering the requests of all users simultaneously, thereby leading to a higher level of overall satisfaction; however, this is a difficult task because it requires a vast order of magnitude to search through a huge set of users. Therefore, in this study, the proposed combination of hybrid of the genetic algorithm and user-preference algorithm is used to overcome this problem. The results overwhelmingly verify that the proposed hybrid approach outperformed previously known used methods significantly.

An In-Network Replica Selection Framework for Latency-Critical Distributed Data Stores

An In-Network Replica Selection Framework for Latency-Critical Distributed Data Stores

Adaptive Replica Creation and Selection Strategies for Latency-Aware Application in Collaborative Edge-Cloud System

Abstract There are many research problems in cloud replica management such as low data reliability, unbalanced node load and large resource consumption. The strategy and status of replica creation, replica placement and replica selection are analyzed. The replica creation based on access tendency (DRC-AT), the replica placement based on user request response time and storage capacity (DRP-RS) and the replica selection based on response time (DRS-RT) are proposed. The DRC-AT algorithm introduces the two parameters of file popularity and period value of file popularity, calculates the file access tendency periodically and decides the creation and deletion of the replica of the file according to the size of the file access tendency. The DRP-RS algorithm evaluates the user’s request response time and storage capacity to select the best node set to place the replica. The DRS-RT algorithm returns to the user the node with the strongest service capability that contains the user’s requested data. Experiments show that the algorithm can improve the speed of data reading by the client, improve the resource utilization, balance the load of the node and improve the overall performance of the system.

Method for replica selection in the Internet of Things using a hybrid optimisation algorithm

Internet of Things (IoT) as a new technological revolution has been proposed recently wherein the things are connected over the Internet. Because of the inherent characteristics of IoT for storage of data at untrusted and heterogeneous hosts, data replication across large geographic distances for efficient data management is unavoidable. The selection of appropriate replication things in the IoT, which reduces response time and cost is one of the most important issues of data management. Since this problem is an NP-hard problem, classic approaches are not efficient to solve this issue, and evolutionary algorithm such as ant colony optimisation (ACO) and genetic algorithm (GA) seems to be very useful. This study offers a method based on a combination of ACO and a GA to solve this problem. In the proposed method, the ACO has been used to create diversity, and afterwards, the GA is performed to provide a full search over the search space. The obtained results have shown the better performance of the proposed method in comparison with ACO, the High-QoS First-Replication (HQFR), and the Response Time-based Replica Management algorithms with regard to waiting time. In addition, the obtained results have revealed the better performance of the hybrid method in comparison with HQFR and the Dynamic Cost-aware Re-replication and Re-balancing Strategy.

Scalable replica selection based on node service capability for improving data access performance in edge computing environment

The replica strategies in traditional cloud computing often result in excessive resource consumption and long response time. In the edge cloud environment, if the replica node cannot be managed efficiently, it will cause problems such as low user’s access speed and low system fault tolerance. Therefore, this paper proposed replica creation and selection strategy based on the edge cloud architecture. The dynamic replica creation algorithm based on access heat (DRC-AH) and replica selection algorithms based on node service capability (DRS-NSC) were proposed. The DRC-AH uses data block as replication granularity and Grey Markov chain to dynamically adjust the number of replicas. After the replica is created, when client receives the user’s request, the DRS-NSC selects the best replica node to respond to the user. The experiments show that the proposed algorithms have significant advantages in prediction accuracy, user’s request response time, resource utilization, etc., and improve the performance of the system to a certain extent.

Understanding and improvement of the selection of replica servers in key–value stores

In current large-scale distributed key–value stores, the tail latency of the hundreds of key–value access operations generated by an end-user request determines the response time of this request. Therefore, this tail latency has great impacts on the user experience and revenue. Replica selection algorithms, which select the best replica server for the service of each key–value access operation as much as possible, is the key to cut the tail latency of these key–value access operations. This paper summarizes current replica selection algorithms, including both the algorithms employed by current key–value stores and the classic algorithms of other similar systems. These algorithms are classified into three categories: information-agnostic, client-independence and feedback, according to their demanded information. As a step further, simulation-based performance analysis of these algorithms is conducted. The result brings us the insights that the response time (RPT) is useful to measure the service rate, but will lead to the herd behaviors. Moreover, the number of outstanding key–value access operations (OSKs) is helpful to both the selection of the fastest replica server and the avoidance of herd behaviors. Based on these insights, we design the L2 algorithm by assembling the basic ideas of the Least OSK algorithm and the Least RPT algorithm. The L2 algorithm is much simpler than the recently proposed C3 algorithm and has a similar best performance with C3 as confirmed by the simulation and experimental results.

Haste makes waste: The On–Off algorithm for replica selection in key–value stores

In current large-scale distributed key–value stores, the tail latency of the key–value accesses generated by end-user requests is crucial to the response time of these requests. To cut the tail latency, the replica selection algorithm, which selects a replica server for each key by a client and thus determines the latency of each key–value access, is crucial. Naturally, all of current replica selection algorithms send keys out immediately to reduce the tail latencies of key–value accesses. In this paper, we find that sending out keys in haste makes waste of the chance to select a better replica server certain time later, and suggest to await for a better replica server available when all current replica servers are bad. To realize this idea, we develop the On–Off algorithm, which recognizes bad replica servers according to the feedback information and put them into the OFF state. Special attention is paid on the time interval replica servers stay at the OFF state. Obviously, the On–Off algorithm adds awaiting time at clients, but can greatly reduce the dominated queuing delays at replica servers. In total, the On–Off algorithm improves the 99th latency by about 29% under the default simulation configuration and outperforms the C3 algorithm proposed recently under kinds of scenarios.