Data Replication

In data grid, using reservation is accepted to provide scheduling and service quality. Users need to have an access to the stored data in geographical environment, which can be solved by using replication, and an action taken to reach certainty. As a result, users are directed toward the nearest version to access information. The most important point is to know in which sites and distributed system the produced versions are located. By selecting a suitable place for versions, the versions having performance, efficiency and lower access time are used. In this study, an efficient method is presented to select the best place for those versions created in data grid by using the users’ firefly algorithm which is compared with two algorithms. Results show that firefly algorithm has better performance than others. Index Terms data grid, data replication, firefly algorithm.

Data grids deal with a huge amount of data regularly. It is a fundamental challenge to ensure efficient accesses to such widely distributed data sets. Creating replicas to a suitable site by data replication strategy can increase the system performance. It shortens the data access time and reduces bandwidth consumption. In this paper, a dynamic data replication mechanism called Latest Access Largest Weight (LALW) is proposed. LALW selects a popular file for replication and calculates a suitable number of copies and grid sites for replication. By associating a different weight to each historical data access record, the importance of each record is differentiated. A more recent data access record has a larger weight. It indicates that the record is more pertinent to the current situation of data access. A Grid simulator, OptorSim, is used to evaluate the performance of this dynamic replication strategy. The simulation results show that LALW successfully increases the effective network usage. It means that the LALW replication strategy can find out a popular file and replicates it to a suitable site without increasing the network burden too much.

Eventual consistency is demanded nowadays in geo-replicated services that need to be highly scalable and available. According to the CAP constraints, when network partitions may arise, a distributed service should choose between being strongly consistent or being highly available. Since scalable services should be available, a relaxed consistency (while the network is partitioned) is the preferred choice. Eventual consistency is not a common data-centric consistency model, but only a state convergence condition to be added to a relaxed consistency model. There are still several aspects of eventual consistency that have not been analysed in depth in previous works: 1. which are the oldest replication proposals providing eventual consistency, 2. which replica consistency models provide the best basis for building eventually consistent services, 3. which mechanisms should be considered for implementing an eventually consistent service, and 4. which are the best combinations of those mechanisms for achieving different concrete goals. This paper provides some notes on these important topics.

Modern text retrieval systems often provide a similarity search utility, that allows the user to find efficiently a fixed number h of documents in the data set that are the most similar to a given query (here a query is either a simple sequence of keywords or a full document). We consider the case of a textual database made of semi-structured documents. For example, in a corpus of bibliographic records any record may be structured into three fields: title, authors and abstract, where each field is an unstructured free text. Each field, in turns, may be modelled with a specific vector space. The problem is more complex when we also allow users to associate at query time to each vector space a weight influencing its contribution to the overall dynamic aggregated and weighted similarity. We investigate the use of metric k-center clustering to prune the search space at query time. The embedding of the weights in the data structure is investigated with the purpose of allowing users query customization without any data replication. The validity of our approach is demonstrated experimentally by showing significant quality/time performance improvements over two state of the art methods. We also speed up the pre-processing time by a factor at least thirty with respect to a method based on k-means clustering.

Dans un contexte d'utilisation de ressources hétérogènes, la performance reste le critère traditionnel pour la planification de capacité. Mais, de nos jours, tenir compte de la variable énergétique est devenu une nécessité. Cet article s'attaque au problème de l'efficacité énergétique pour la répartition de charge dans les systèmes distribués. Nous proposons une gestion efficace en énergie des ressources par l'ajout de fonctionnalités de gestion des événements liés à l'énergie, selon des règles définies par l'utilisateur. Nous implémentons ces fonctionnalités au sein de l'intergiciel DIET, qui permet de gérer la répartition de charge afin de mettre en évidence le coût des compromis entre la performance et la consommation d'énergie. Notre solution et son intérêt sont validés au travers d'expériences en évaluant la performance et la consommation électrique mettant en concurrence trois politiques d'ordonnancement. Nous mettons en avant le gain obtenu en terme énergétique tout en essayant de minimiser les écarts de performance. Nous offrons également à l'intergiciel responsable de l'ordonnancement une réactivité face aux variations énergétiques. Mots-clés : Systèmes distribués, efficacité énergétique, répartition de charge, planification de capacité 1. Le classement du TOP500.

Big Data applications allow to successfully analyze large amount of data not necessarily structured, though at the same time they present new challenges. For example, predicting the performance of frameworks such as Hadoop can be a costly task, hence the necessity to provide models that can be a valuable support for designers and developers. This paper provides a new contribution in studying a new modeling approach based on fluid Petri nets to envision MapReduce jobs execution time.

The identification of the highest yielding cultivar for a specific environment on the basis of both genotype (G) and genotype × environment (GE) interaction would be useful to breeders and producers since yield estimates based only on G and environment (E) effects are insufficient. The objective of this study was to demonstrate the usefulness of additive main effects and multiplicative interactions (AMMI) model analysis and G plus GE interaction (GGE) biplots, obtained from sites regression (SREG) model analysis in interpreting GE grain yield data. Replicated grain yield data of six rice (Oryza sativa L.) cultivars (Cocodrie, Cypress, Jefferson, Lemont, Saber, and Wells) from three main cropping seasons (2000, 2001, and 2002) at four locations in Texas, USA (Bay City, Eagle Lake, Ganado, and Beaumont) were obtained and used for this purpose. Through AMMI model analysis, the magnitude and significance of the effects of GE interaction and its interaction principal components relative ...

The emergence and widespread adoption of Grid computing has been fueled by continued growth in both our understanding of application requirements and the sophistication of the technologies used to meet these requirements. We provide an introduction to Grid applications and technologies and discuss the important role that resource management will play in future developments. ¢ 03], released in 2003. Building on and significantly extending GT2 concepts and technologies, OGSA firmly aligns Grid computing with broad industry initiatives in service-oriented architecture and Web services. In addition to defining a core set of standard interfaces and behaviors that address many of the technical challenges introduced above, OGSA provides a framework within which can be defined a wide range of interoperable, portable services. OGSA thus provides a foundation on which can be constructed a rich Grid technology ecosystem comprising multiple technology providers. Thus, we see, for example, major efforts under way within the U.K. eScience programme to develop data access and integration services [ACK ¢ 04, PAD ¢ 02]. Concurrent with these developments we see a growing recognition that largescale development and deployment of Grid technologies is critical to future success in a wide range of disciplines in science, engineering, and the humanities [ADF ¢ 03], and increasingly large investments within industry in related The Grid in a Nutshell

Trying to remember something now typically improves your ability to remember it later. However, after watching a video of a simulated bank robbery, participants who verbally described the robber were 25% worse at identifying the robber in a lineup than were participants who instead listed U.S. states and capitals—this has been termed the “verbal overshadowing” effect (Schooler & Engstler-Schooler, 1990). More recent studies suggested that this effect might be substantially smaller than first reported. Given uncertainty about the effect size, the influence of this finding in the memory literature, and its practical importance for police procedures, we conducted two collections of preregistered direct replications (RRR1 and RRR2) that differed only in the order of the description task and a filler task. In RRR1, when the description task immediately followed the robbery, participants who provided a description were 4% less likely to select the robber than were those in the control con...

The goal of the Globe project is to design and build a middleware platform that facilitates the development of large-scale distributed applications, such as those found on the Internet. To demonstrate the feasibility of our design and to test our ideas, we are currently building a new Internet application: The Globe Distribution Network. The Globe Distribution Network, or GDN, is an application for the efficient, worldwide distribution of free software and other free data. The GDN can be seen as an improvement to anonymous FTP and the World Wide Web due to its flexibility and extensive support for replication. This paper describes the design of the GDN. We start by explaining how the replication facilities of the Globe middleware are used to make the GDN efficient, and how these facilities are implemented. Next, we present the architecture of the GDN and discuss how the Domain Name System can be used as a first approach towards a worldwide service for naming software packages and other entities. This is followed by an analysis of the security requirements for the GDN and measures taken to satisfy these requirements. We hope to make Globe and GDN itself available for free under the BSD license by 2001.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.  A new approach for the Entity Matching task parallelization is proposed.  The idea relies on performing a MapReduce-based multi-pass adaptive blocking strategy.  The proposed approach shows significant superior performance efficiency.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.  A new approach for the Entity Matching task parallelization is proposed.  The idea relies on performing a MapReduce-based multi-pass adaptive blocking strategy.  The proposed approach shows significant superior performance efficiency.

Databases have become a crucial component in modern information systems. At the same time, they have become the main bottleneck in most systems. Database replication protocols have been proposed to solve the scalability problem by scaling out in a cluster of sites. Current techniques have attained some degree of scalability, however there are two main limitations to existing approaches. Firstly, most solutions adopt a full replication model where all sites store a full copy of the database. The coordination overhead imposed by keeping all replicas consistent allows such approaches to achieve only medium scalabilitiy. Secondly, most replication protocols rely on the traditional consistency criterion, 1-copy-serializability, which limits concurrency, and thus scalability of the system. In this paper, we first analyze analytically the performance gains that can be achieved by various partial replication configurations, i.e., configurations where not all sites store all data. From there, we derive a partial replication protocol that provides 1-copy-snapshot isolation as correctness criterion. We have evaluated the protocol with TPC-W and the results show better scalability than full replication.

In recent years collaborative editing systems such as wikis, GoogleDocs and version control systems became very popular. In order to improve reliability, fault-tolerance and availability shared data is replicated in these systems. User misbehaviors can make the system inconsistent or bring corrupted updates to replicated data. Solutions to secure data history of state-based replication exist, however they are hardly applied to operation-based replication. In this paper we propose an approach to secure log in operation-based ...

Spatial data analysis applications are emerging from a wide range of domains such as building information management, environmental assessments and medical imaging. Time-consuming computational geometry algorithms make these applications slow, even for medium-sized datasets. At the same time, there is a rapid expansion in available processing cores, through multicore machines and Cloud computing. The confluence of these trends demands effective parallelization of spatial query processing. Unfortunately, traditional parallel spatial databases are ill-equipped to deal with the performance heterogeneity that is common in the Cloud. We introduce Niharika, a parallel spatial data analysis infrastructure that exploits all available cores in a heterogeneous cluster. Niharika first uses a declustering technique that creates balanced spatial partitions. Then, Niharika adapts to performance heterogeneity and processing skew in the spatial dataset using dynamic loadbalancing. We evaluate Niharika with three load-balancing algorithms and two different spatial datasets (both from TIGER) using Amazon EC2 instances. Niharika adapts to the performance heterogeneity in the EC2 nodes, thereby achieving excellent speedups (e.g., 63.6X using 64 cores on 16 4-core EC2 nodes, in the best case) and outperforming an approach that does not adapt.

This report presents the V7 milestone of the Phase-Locked Quantum-Plasma Processor. We confirm that localised phase nodes can be copied between heterogeneous fields while conserving functionality. Experiments cover 2-D and 3-D replication, behaviour reuse, multi-agent crowding dynamics, and goal reorientation (task-shift) without any retraining or symbolic logic. All GIF animations referenced below are available in the Simulations V7.zip archive.

The increasing volume and complexity of Big Data have led to the development of distributed processing frameworks such as Apache Spark, particularly its Python interface, PySpark, which allows for large-scale data processing in cloud environments. This paper investigates the optimization of data locality and caching strategies to improve the performance and scalability of Big Data workflows running on Google Cloud Platform (GCP). Through a series of experiments, various configurations of data placement, replication, and caching techniques-both in-memory and disk-based-were evaluated for their impact on key performance metrics, including execution time, latency, and throughput. The study also assesses the scalability of workflows as data sizes increase, identifying the configurations that allow PySpark workflows to handle growing datasets efficiently. The results reveal that optimized data locality, combined with well-tuned caching strategies, can significantly improve performance and scalability, offering a pathway for businesses to enhance their cloud-based Big Data systems. Furthermore, the findings provide valuable insights for organizations seeking to reduce costs, accelerate decisionmaking, and improve the efficiency of their data processing workflows. This paper contributes to the ongoing efforts to optimize distributed Big Data processing frameworks in cloud environments and offers practical guidelines for configuring PySpark workflows for maximum performance.

In this paper, we address present an optimal solution for the problem of multimedia object placement for hybrid transparent data replication. The performance objective is to minimize the total access cost by considering both transmission cost and transcoding cost. The performance of the proposed solution is evaluated with a set of carefully designed simulation experiments for various performance metrics over a wide range of system parameters. The simulation results show that our solution consistently and significantly outperforms comparison solutions in terms of all the performance metrics considered.

Scheduling in traditional distributed systems has been mainly studied for system performance parameters without data transmission requirements. With the emergence of Data Grids (DGs) and Data Centers, data-aware scheduling has become a major research issue. In this work we present two implementations of classical genetic-based data-aware schedulers of independent tasks submitted to the grid environment. The results of a simple empirical analysis confirm the high effectiveness of the genetic algorithms in solving very complex data intensive combinatorial optimization problems.

Data replication is a promising technique for increasing access performance and data availability in Data Grid (DG) systems. Current work on data replication in Grid systems focuses on infrastructure for replication and mechanisms for creating or deleting replicas. The important problem of determining minimum number of replicas, as well as their locations in DG has not been well studied. In this paper, we propose an algorithm is formulated by using dynamic programming-based algorithm to find Optimal Number of Replicas (ONR) of a dataset over DG systems, such that the read cost (i.e. transferring object over the Data Grid systems to the end-user)plus the cost of storage (i.e. site building cost) is minimized. We have also proposed a sketch of the proof for our algorithm and its integrity.

This paper advances the study of democratic trajectorieswhether democracies deepen, stagnate, erode or break down over time. We show that econometric panel models usually neglect cumulative effects, which are implicitly central to many theories of democratic change. Some important factors, such as economic growth, have cumulative effects that shape medium-to long-term regime trajectories. To overcome the limitations of conventional statistical estimators, we propose the use of latent growth curve models, which are better able to capture cumulative processes. We demonstrate the advantages of this approach by analyzing the trajectory of 103 democratic regimes inaugurated after 1974. Conventional estimators fail to predict democratic trajectories, while latent growth curve models properly capture cumulative effects.

Alhamdulillah, thanks God, without H~m I am nothing. I would like to express my sillcere thanks to my major adviser Dr. Mitchell L. Neilsen for his guidanceẽ ncouragement, help, and support for the completion of my thesis. Without his constant support and inspiring ideas, this thesis would have been impossible. I also would like to thank Dr. K.M. George and Dr. H. Lu for serving on my graduate committee and for their precious time spending on my thesis. I also want to thank Dr. J.P. Chandler for his generous support a.nd help. I would also like to thank to my Indonesian Government represented by The Agency for The Assessment and Application of Technology (BPP Teknologi) for giving me an opportunity to pursue my education and sponsoring me. To Nusantara Aircraft Industries Limited (PT IPTN) where I have been working for, I want to thank for aU support that has been given to me and for giving me an opportunity to get my dream pursuing to the higher education. Last~but certainly not least, I wish to express with the deepest of my h art my sincere thanks to my parents, the ones who always instil in me the important of learning and always pray to God for me. My thanks also go to all of my friends who have helped, encouraged, and advised me during my stay in Stillwater.

Trying to remember something now typically improves your ability to remember it later. However, after watching a video of a simulated bank robbery, participants who verbally described the robber were 25% worse at identifying the robber in a lineup than were participants who instead listed U.S. states and capitals—this has been termed the “verbal overshadowing” effect (Schooler & Engstler-Schooler, 1990). More recent studies suggested that this effect might be substantially smaller than first reported. Given uncertainty about the effect size, the influence of this finding in the memory literature, and its practical importance for police procedures, we conducted two collections of preregistered direct replications (RRR1 and RRR2) that differed only in the order of the description task and a filler task. In RRR1, when the description task immediately followed the robbery, participants who provided a description were 4% less likely to select the robber than were those in the control con...

In the second section we discuss in general terms how the Varieties of Democracy (V-Dem) project differs from extant indices and how the novel approach taken by V-Dem might assist the work of activists, professionals, and scholars. Appendix A addresses the possible uses of V-Dem for program evaluation, Appendix B offers a glossary of key terms used in this document, and Appendix C clarifies search terms used for several analyses in Table . Many attempts have been made to measure democracy. It is somewhat complicated to identify these indices because they do not always refer explicitly to democracy. Nonetheless, we include an index in our survey if it is commonly viewed as representing democracy or some aspect of democracy, and if it features fairly broad country coverage. This includes the BNR index developed by Bernhard, Nordstrom & Reenock (2001); the Bertelsmann Transformation Index ("BTI") directed by the Bertelsmann Stiftung (various years); the Democracy Barometer developed by Wolfgang Merkel & associates (Bühlmann, Merkel, Müller & Weßels 2012); the BMR index developed by Boix, Miller & Rosato (2013), the Contestation and Inclusiveness indices developed by Coppedge, Alvarez, & Maldonado (2008); the Political Rights, Civil Liberty, Nations in Transit, and Countries at the Crossroads indices, all sponsored by Freedom House (freedomhouse.org); the Economist Intelligence Unit (2010) index ("EIU"); the Unified Democracy Scores ("UDS") developed by Pemstein, Meserve & Melton (2010); the Polity2 index from the Polity IV database (Marshall, Gurr & Jaggers 2014); the democracy-dictatorship ("DD") index developed by Adam Przeworski & colleagues (Alvarez, Cheibub, Limongi & Przeworski 1996; Cheibub, Gandhi & Vreeland 2010); the Lexical index of electoral democracy developed by Skaaning, Gerring & Bartusevičius (2015); the Competition and Participation indices developed by Tatu Vanhanen (2000); and the Voice and Accountability index developed as part of the Worldwide Governance Indicators ("WGI") (Kaufmann, Kraay & Mastruzzi 2010). 2 In order to make comparisons in a systematic fashion we summarize key features of 2 Other indices, not included in Table , may be briefly listed: the Political Regime Change [PRC] dataset

Patients with schizophrenia frequently demonstrate hypofrontality in tasks that require executive processing; however questions still remain as to whether prefrontal cortex dysfunctions are specific to schizophrenia, or a general feature of major psychopathology. Context processing is conceptualized as an executive function associated with attention and working memory processes. Impairment in the ability of patients with schizophrenia to represent and maintain context information has been previously reported in a number of studies. To examine the question of the specificity of a context processing deficit to schizophrenia, we used functional MRI and an expectancy AX continuous performance task designed to assess context processing in a group of healthy controls (n = 9), depressed patient controls (n = 10), and patients with schizophrenia (n = 7). The behavioral performance was consistent with a context processing deficit in patients with schizophrenia, but not those with depression. The imaging data replicate previous results in showing abnormal activity in the right middle frontal gyrus (BA9) in schizophrenia patients related to context processing.

To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. In this paper, we propose lazy replication as a way to preserve consistency by exploiting the semantics of the service's operations to relax the constraints on ordering. Three kinds of operations are supported: operations for which the clients define the required order dynamically during the execution. operations for which the service defines the order, and operations that must be globally ordered with respect to both client ordered and service ordered operations. The method performs well in terms of response time, amount of stored state, number of messages, and availability. It is especially well suited to applications in which most operations require only the client-defined order.

A real time distributed computing has heterogeneously networked computers to solve a single problem. So coordination of activities among computers is a complex task and deadlines make more complex. The performance of the system depends on many factors such as traffic workloads data base system architecture, underlying processor, disk speeds, concurrency control, transaction management etc.[1,2,3,4,5,6].  A simulation study have  to be performed to analyze the performance under different transaction scheduling, different workloads, arrival rate priority policies, altering slack factors and preemptive policies. The performance of the distributed system under various conditions is to be monitored and parameters such as arrival rate, transaction size, transaction distribution policies, and execution time are to be analyzed

Purpose: Genome-wide association studies (GWAS) have identified 6q25, which incorporates the oestrogen receptor a gene (ESR1), as a quantitative trait locus for areal bone mineral density (BMD a ) of the hip and lumbar spine. The aim of this study was to determine the influence of this locus on other bone health outcomes; calcaneal ultrasound (QUS) parameters, radial peripheral quantitative computed tomography (pQCT) parameters and markers of bone turnover in a population sample of European men. Methods: Eight single nucleotide polymorphisms (SNP) in the 6q25 locus were genotyped in men aged 40-79 years from 7 European countries, participating in the European Male Ageing Study (EMAS). The associations between SNPs and measured bone parameters were tested under an additive genetic model adjusting for centre using linear regression. Results: 2468 men, mean (SD) aged 59.9 (11.1) years had QUS measurements performed and bone turnover marker levels measured. A subset of 628 men had DXA and pQCT measurements. Multiple independent SNPs showed significant associations with BMD using all three measurement techniques. Most notably, rs1999805 was associated with a 0.10 SD (95%CI 0.05, 0.16; p = 0.0001) lower estimated BMD at the calcaneus, a 0.14 SD (95%CI 0.05, 0.24; p = 0.004) lower total hip BMD a , a 0.12 SD (95%CI 0.02, 0.23; p = 0.026) lower lumbar spine BMD a and a 0.18 SD (95%CI 0.06, 0.29; p = 0.003) lower trabecular BMD at the distal radius for each copy of the minor allele. There was no association with serum levels of bone turnover markers and a single SNP which was associated with cortical density was also associated with cortical BMC and thickness. Conclusions: Our data replicate previous associations found between SNPs in the 6q25 locus and BMD a at the hip and extend these data to include associations with calcaneal ultrasound parameters and radial volumetric BMD.

Cloud computing is one of the upcoming latest new computing paradigm where applications and data services are provided over the Internet. Cloud computing is attractive to business owners as it eliminates the requirement for users to plan ahead for provisioning, and allows enterprises to start from the small and increase resources only when there is a rise in service demand. While dealing with cloud computing, a number of issues are confronted like heavy load or traffic while computation. Job scheduling is one of the answers to these issues. The main advantage of job scheduling algorithm is to achieve a high performance computing and the best system throughput. Job scheduling is most important task in cloud computing environment because user have to pay for resources used based upon time. This paper aims to do the insight study of various existing job scheduling algorithms and proposes a hybrid job scheduling algorithm for enhancing the performance in cloud computing environment.

Reproducibility is a defining feature of science, but the extent to which it characterizes current research is unknown. We conducted replications of 100 experimental and correlational studies published in three psychology journals using high-powered designs and original materials when available. Replication effects were half the magnitude of original effects, representing a substantial decline. Ninety-seven percent of original studies had statistically significant results. Thirty-six percent of replications had statistically significant results; 47% of original effect sizes were in the 95% confidence interval of the replication effect size; 39% of effects were subjectively rated to have replicated the original result; and if no bias in original results is assumed, combining original and replication results left 68% with statistically significant effects. Correlational tests suggest that replication success was better predicted by the strength of original evidence than by characteris...

Nous sommes à l'aube d'une énorme explosion de données et la quantité à traiter par les entreprises est de plus en plus grande. Pour faire face à ce chalenge, Google a développé MapReduce, un modèle de programmation parallèle qui est en train de devenir l'outil de facto pour l'analyse des systèmes Big Data. Bien que dans une certaine mesure son utilisation est déjà très répandue dans l'industrie, garantir les performances d'un système aussi complexe pose de grands problèmes et sa gestion nécessite un haut niveau d'expertise. Cet article répond à ces défis en proposant le premier système autonome qui garantit des contraintes de temps de réponse pour une charge de travail MapReduce simultanée. Nous développons le premier modèle dynamique d'une grappe MapReduce. De plus, un contrôle en boucle fermée est conçu et implémenté pour garantir un temps de réponse donné. Un contrôle d'anticipation de type "feedforward" est également rajouté pour améliorer la réponse du système en présence de perturbations, en l'occurrence, la variation du nombre de clients. L'approche est validée en ligne sur une grappe MapReduce avec 40 noeuds utilisant une charge de travail intensive de type Business Intelligence. Nos expériences montrent que le contrôle ainsi conçu peut garantir les contraintes de temps de réponse.

Multi-tier architectures provide a means for building scalable distributed services. Caching is a classical technique for enhancing the performance of systems (e.g. database servers, or web servers). Although caching solutions have been successfully studied for individual tiers of multi-tier systems, if collectively applied, these solutions may violate the coherence of cached data. This paper precisely addresses this issue. It presents e-Caching, an end-to-end caching system for multi-tier services aimed at guaranteeing data consistency across caches. The paper describes a novel caching protocol to automatically detect data dependencies and apply inter-cache invalidations. The implemented protocol is evaluated with the TPC-W online multi-tier bookstore service. The evaluation shows that e-Caching improves service performance compared to regular caching techniques while guaranteeing global data consistency in multi-tier systems. Today, many applications are implemented using multi-tier architectures. Internet services, which range from Web servers to streaming media services, are based on this kind of architectures. Classical three-tier Web applications start with requests from Web clients that flow through an HTTP front-end server and provider of static content, then to an enterprise server to execute the business logic of the application and generate web pages on-the-fly, and finally to a database that stores non-ephemeral data. To face high loads in multi-tier systems, a commonly used approach is caching . Roughly speaking, caching consists in making a copy of request results for faster future reuse of these results. Current caching solutions may successfully apply at individual tiers of multi-tier systems to cache database query results , application business objects , or dynamic Web documents . Intuitively, caching back-end tier results (i.e. database query results) provides higher cache hit ratio than caching front-end tier query results, since the same back-end query may be issued and shared by different front-end queries. Whereas, caching front-end tier results offers a greater benefit in terms of service response times in case of a cache hit, since front-end caching saves time on the front-end and on the back-end. Therefore, caching solutions at these different levels are complementary and would improve the overall performance of multi-tier systems. However, if put all together, existing individual caching solutions of multi-tier systems might result in globally inconsistent cached data across tiers, in case of dynamic and inter-dependent data. The objective of this paper is to precisely bridge the gap between performance and consistency of caching in multi-tier systems as follows : -Automatic data analysis to detect inter-dependent data across caches of all tiers. -Data consistency across multi-tier cache systems by using the generated data dependencies. -Transparent caching for an easier and automated integration of caching to multi-tier systems. In this paper, we present e-Caching, an end-to-end caching system that provides consistency across cache levels. It guarantees consistency of cached data through an automatic multi-tier flow analysis. We describe a novel caching protocol that guarantees consistency of cached data through multi-tier flow

Sens}@lip6.fr Résumé Le Cloud computing marque une nouvelle avancée vers l'infrastructure informatique dématérialisée. Le Cloud fournit des ressources informatiques, logicielles ou matérielles, accessible à distance en tant que service. L'adoption de ce modèle soulève un certain nombre de défis, notamment au sujet de la qualité de service (QoS) des services fournis. En effet, alors qu'on assiste à une multiplication de services similaires sur le Cloud (stockage, environnement de développement, bureautique, etc.), rien ne nous permet d'évaluer la qualité d'un service rendu. Cet article présente une solution pour intégrer la QoS et le contrat SLA (Service Level Agreement) comme éléments à part entière du Cloud. Les contributions de cet article sont les suivantes. Premièrement, nous proposons le modèle de Cloud SLAaaS (SLA aware as a Service) qui apporte une dimension supplémentaire au paradigme de Cloud Computing. Deuxièmement, nous présentons le langage Cloud Service Level Agreement (CSLA) qui permet la définition et l'établissement d'un contrat qui supporte l'instabilité de la QoS dans un environnement hautement dynamique comme le Cloud en gérant finement la violation. Enfin, nous développons deux stratégies dirigées par CSLA pour pouvoir contrôler dynamiquement et de manière autonome la taille et la forme du Cloud pour répondre aux contraintes de SLA : i) une stratégie de planification de capacité pour le redimensionnement sur mesure et ii) une stratégie de verrouillage d'exclusion mutuelle pour assurer un accès exclusif à des ressources partagées sur le Cloud. Cloud Computing, qualité de service (QoS), contrat SLA (Service Level Agreement), planification de capacité, exclusion mutuelle ComPAS'2013 ( RenPar/ SympA/ CFSE

Cloud providers aim to maximise their profits while satisfying tenant requirements, e.g., performance. The relational database management systems face many obstacles in achieving this goal. Therefore, the use of NoSQL databases becomes necessary when dealing with heterogeneous workloads and voluminous data. In this context, we propose a new data replication strategy that balances the workload of nodes and dynamically adjusts the number of replicas while the provider profit is taken into account. Result analysis shows that the proposed strategy reduces the resource consumption, which improves the provider profit while satisfying the tenant performance requirement.

With the rapid growth of emerging applications like social network, semantic web, sensor networks and LBS (Location Based Service) applications, a variety of data to be processed continues to witness a quick increase. Effective management and processing of large-scale data poses an interesting but critical challenge. Recently, big data has attracted a lot of attention from academia, industry as well as government. This paper introduces several big data processing techniques from system and application aspects. First, from the view of cloud data management and big data processing mechanisms, we present the key issues of big data processing, including definition of big data, big data management platform, big data service models, distributed file system, data storage, data virtualization platform and distributed applications. Following the Map Reduce parallel processing framework, we introduce some MapReduce optimization strategies reported in the literature. Finally, we discuss the open issues and challenges, and deeply explore the research directions in the future on big data processing in cloud computing environments.

The increasing adoption of multi-cloud database systems has transformed enterprise data management, enabling enhanced scalability, reliability, and cost efficiency. However, managing databases across multiple cloud providers introduces significant challenges, including data fragmentation, latency, security vulnerabilities, and inconsistencies in synchronization. Traditional approaches to database management struggle to provide seamless interoperability, fault tolerance, and resilience against failures, necessitating innovative architectural solutions.This paper explores the design and implementation of resilient multi-cloud database systems, integrating Distributed Ledger Technology (DLT) for enhanced data integrity, fault tolerance mechanisms to ensure high availability, and cross-platform synchronization techniques for maintaining consistency across heterogeneous cloud environments. DLT, particularly blockchain, offers a decentralized approach to data validation, reducing the risk of tampering and unauthorized modifications while enabling transparent and auditable transactions. Fault tolerance strategies, including redundancy, self-healing systems, and predictive analytics, play a crucial role in mitigating system failures and ensuring business continuity. Additionally, cross-platform synchronization mechanisms, such as conflict-free replicated data types (CRDTs) and real-time consistency protocols, are explored to address latency and data consistency challenges across cloud infrastructures. By integrating these technologies, organizations can enhance the resilience, security, and operational efficiency of multi-cloud database architectures. This paper provides a comprehensive framework for implementing adaptive database management solutions, leveraging AI-driven automation, blockchain-based security, and advanced fault recovery models. The findings highlight best practices for enterprises aiming to achieve scalable, reliable, and fault-tolerant multi-cloud database environments. Future research directions include the role of edge computing in multi-cloud synchronization, quantum-safe cryptographic techniques for DLT security, and AI-driven predictive failure management in cloud-native databases.

In a replicated environment, same data are often distributed to several sites in order to improve the data availability, fault tolerance and faster access. However, when a replica is modified by a user, the other replicas become stale. In such situation a mechanism is needed to keep them consistent, which means all the replicated data have the same content. This can be achieved using a synchronization mechanism and it provides the same view of all the replicas by propagating the modifications done on the first replica to others. This process is usually called as update propagation which is the main part of the consistency mechanism. When replicas are read-only or when no consistency guarantees are required by the users of the replicated system, a synchronization system is not needed. In the present chapter, different replica synchronization techniques are studied. Also, based on the agent based framework proposed in the previous chapter an update propagation mechanism is proposed to...

This paper explores a game-theoretic model for task allocation in distributed systems, where processors with varying speeds and external load factors are considered strategic players. The goal is to understand the impact of processors' strategic behaviors on workload management and overall system efficiency, focusing on the attainment of a pure strategy Nash Equilibrium (NE). The research rigorously develops a formal mathematical model and validates it through extensive simulations, highlighting how NE ensures stability but may not always yield optimal system performance. The adaptive algorithms for dynamic task allocation are proposed to enhance efficiency in real-time processing environments. Results demonstrate that while NE provides stability, the adoption of optimal cooperative strategies significantly improves operational efficiency and reduces transaction costs. The findings contribute valuable insights into the strategic interactions within computational frameworks, offering guidelines for developing more efficient systems. This study not only advances the theoretical understanding of strategic task allocation but also has practical implications for system design and policy-making in areas such as cloud computing and traffic management.

The television industry has undergone a massive transformation with the rise of Internet Protocol Television (IPTV). Unlike traditional cable and satellite TV, IPTV leverages the internet to deliver content, offering flexibility, interactivity, and enhanced viewing experiences. This paper explores the working mechanism of IPTV, its technological foundations, and its potential future in the rapidly evolving digital landscape

In a sensor network information from multiple nodes must usually be aggregated in order to accomplish a certain task. A natural way to view this information gathering is in terms of interactions between nodes that are producers of information, e.g., those that have collected data, detected events, etc., a nd nodes that are consumers of information, i.e., nodes that seek data of certain types. Our overall goal in this paper is to construct efficient schemes allowing consumer and producer nodes to discover each other so that the desired information can be sent quickly to those who seek it. Here, efficiency is an issue for both the producers (limiting the redundancy of where information is stored) as well as the consumers (keeping the query time low). We introduce the notion of distance-sensitive information brokerage and provide schemes for efficiently bringing together information producers and consumers at a cost proportional to the separation between them—even though the consumers ...

The demand for replicability of behavioral results across laboratories is viewed as a burden in behavior genetics. We demonstrate how it can become an asset offering a quantitative criterion that guides the design of better ways to describe behavior. Passing the high benchmark dictated by the replicability demand requires less stressful and less restraining experimental setups, less noisy data, individually customized cutoff points between the building blocks of movement, and less variable yet discriminative dynamic representations that would capture more faithfully the nature of the behavior, unmasking similarities and differences and revealing novel animal-centered measures. Here we review ten tools that enhance replicability without compromising discrimination. While we demonstrate the usefulness of these tools in the context of inbred mouse exploratory behavior they can readily be used in any study involving a high-resolution analysis of spatial behavior. Viewing replicability as a design concept and using the ten methodological improvements may prove useful in many fields not necessarily related to spatial behavior.

The issue of data replication is considered in the context of a restricted system model motivated by certain distributed data-warehousing applications. A new replica management protocol is defined for this model in which gIobaI serializability is ensured, while message overhead and deadlock frequency are less than in previously published work. The advantages of the protocol arise from its use of a lazy approach to update of secondary copies of replicated data and the use of a new concept, virtual sites, to reduce the potential for conflict among global transactions.

This study was conducted to evaluate the significance and magnitude of the effect of genotype x environment (GE) interaction on corn grain yield, and to determine the best genotype for each major corn-growing region in the Philippines. Replicated grain yield data of corn hybrids common across locations for each time period (consisting of two consecutive seasons) from 1996 to 2003 were examined and interpreted through the additive main effects and multiplicative interaction (AMMI) model and biplot analyses. The nominal yield (expected yield without the location main effect) of each genotype at each environment (location-season combination) was computed. Mega-environments (group of environments within each period that had the same highest nominal yielding genotype) were identified. Genotype (G), environment (E) and GE interaction were significant in all 14 time periods and accounted for an average of 5.3%, 83.8% and 10.5% of the variation in grain yield, respectively. The stability (range in nominal yield across environments) of each of the 55 genotypes (36 cultivars recommended by National Seed Industry Council [NSIC] and 19 non-recommended cultivars) that were identified as the highest nominal yielding genotype for their respective mega-environment was estimated. The highest nominal yielding genotype for the dry season (DS) was different from that for the wet season (WS) in 65 of the 95 location-period combinations. This list would be useful in recommending the appropriate genotype to grow in a particular environment as a cultivar for production or as a check cultivar in breeding or yield trials. It was estimated that nominal yields during the DS and WS would be 7.44 and 7.46 t ha-1, respectively, if the highest nominal yielding genotypes are selected and grown specifically in their respective mega-environments. For the 2002WS to the 2003DS time period, which included 14 environments, six mega-environments and their respective highest nominal yielding genotype were identified, CPX818 being the best performer based on its high and stable nominal yield. This study showed that GE interaction should be considered when estimating the yields of genotypes at different locations or when recommending the best genotype for the major corn- growing regions in the Philippines.

Rice breeders consider grain yield and milled rice percentages in developing cultivars, but usually do not consider gross income. This study’s objectives were to identify rice genotypes that produced high and stable expected gross incomes using genotype plus genotype × environment (GGE) biplot analysis. Uniform Regional Rice Nursery data on 47 long-grain genotypes grown at five locations (AR, LA, MO, MS, and TX) during 2001 to 2003 were analyzed. Gross income of each genotype was estimated based on grain yield, milled rice percentages, market prices, and direct and counter-cyclical payments. Based on GGE biplot analysis, the genotypes with the highest yield and highest gross income for the main crop were different in 12 out of 13 environments. RU0103184, Francis, and RU0003178 were the genotypes with the highest gross income in six, four, and three environments, respectively. Rice breeders should consider gross income as a selection criterion in the release of new cultivars. Key words: Rice, GGE, GE, breeding, gross income

The Mariposa distributed data manager uses an economic model for managing the allocation of both storage objects and queries to servers. In this paper, we present extensions to the economic model which support replica management, as well as our mechanisms for propagating updates among replicas. We show h o w our replica control mechanism can be used to provide consistent, although potentially stale, views of data across many machines without expensive per-transaction synchronization. We present a rule-based conict resolution mechanism, which can be used to enhance traditional time-stamp serialization. We discuss the eects of our replica system on query processing for both read-only and read-write queries. We further demonstrate how the replication model and mechanisms naturally support name service in Mariposa. optimization and query processing semantics. Finally, w e show h o w our replication system supports name service without additional mechanisms. A Mariposa system consists of a collection of sites which provide storage, query processing, and name service. More than one service can be provided by one site. The goal of Mariposa is to provide the following features: Scalability: Our goal is for Mariposa to scale to 10; 000 sites. This makes it necessary for Mariposa sites to operate autonomously and without global synchronization. Database activities such as class creation, updates, deletions, fragmentation and data movement m ust happen without notifying any central authority. Fragmentation: E v ery Mariposa table (class) is horizontally partitioned into a collection of fragments which together store the instances of the table. The collection of fragments can be structured (partitioned by predicate-based distribution criteria) or unstructured (partitioned randomly or in round-robin order). Data movement: F ragments can move from one site to another without quiescing the database. Data movement makes it possible for Mariposa sites to ooad data objects, resulting in load balancing and better system throughput. Flexible copies: Copies can enhance data availability and provide faster query processing through parallel execution. However, the cost of maintaining the consistency of a large set of replicas can be prohibitive if conventional techniques (e.g., two-phase commit) are used. Mariposa provides a replica-management system that avoids the expensive synchronization requirements of conventional replica systems without sacricing transaction serializability. Copies are at the granularity of fragments. Rule-based system management: The behavior of Mariposa sites is controlled by a production rule system, provided as part of the Rush scripting language [SAH94]. Rules are of the form on event where condition do action. The rule system is intended as a exible tool to implement policy governing buying and selling data, pricing for query processing, etc. In [STON94b], we expressed query processing in terms of the following economic framework. A user presents a query to a broker module. Along with the query, the user species a bid curve. The bid curve i s i n t w o dimensions, cost and delay, and species the maximum payment for a query to be answered within a given delay. Some users will be willing to pay more for a fast answer, resulting in a bid curve with a steep slope. Others will be willing to wait, but have a maximum price they are willing to pay, resulting in a atter bid curve. The broker accepts the query and the

One of the basic services in grids is the transfer of data between remote machines. Files may be transferred at the explicit request of the user or as part of delegated resource management services, such as data replication or job scheduling. GridFTP is an important tool for such data transfers since it builds on the common FTP protocol, has a large user base with multiple implementations, and it uses the GSI security model that allows delegated operations. This paper presents a workload analysis of the implementation of the GridFTP protocol provided by the Globus Toolkit. We studied more than 1.5 years of traces reported from all over the world by Globus GridFTP installed components. Our study focuses on three dimensions: first, it quantifies the volume of data transferred and characterizes user behavior. Second, it attempts to show how tuning capabilities are used in practice. Finally, it quantifies the user base as recorded in the database and highlights the usage trends of this software component.

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.