Franck Morvan

Large Scale Data Management in Grid Systems: a Survey

Today the grid computing, intended initially for the intensive computing, open towards the management of voluminous, heterogeneous, and distributed data on a large-scale environment. The grid data management raises new problems and presents real challenges: resource discovery, efficiency of access, autonomic management, security, and benchmarking. This importance comes out of characteristics offered by grid systems: autonomy, heterogeneity and dynamicity of nodes. Firstly, we recall the fundamental problems of the large scale data management in grid systems and characteristics of these systems. Then, we describe in a highlight way proposed approaches (Web services, P2P techniques, Agent-based approach) for resource discovery. The remainder of the paper is devoted to point out the contributions of mobile agents for some problems of large scale data management, in particularly: dynamic query optimization, task placement, and embedded cost model. We show how mobile agents can help for decentralized control, and scaling.

Mobile Agents Based Self-Adaptive Join for Wide-Area Distributed Query Processing

In this article, optimization of decision support queries is considered in the context of wide-area distributed databases. An original approach based on the “mobile agent†paradigm is proposed and evaluated. Agents’ autonomy and reactivity allow operators of the execution plan to adapt dynamically to estimation errors on relations and to evolutions in the state of the execution system, avoiding time overheads commonly associated with centralized monitoring. We present decentralized self-adaptive algorithms for dynamic optimization of join operators, and their implementations in Java using mobile agents. Then, we evaluate performance depending on error rate on statistical information on database, and on communication bandwidth and CPU frequency. The results show that the agent-based approach can lead to a significant reduction of response time and provide decision criteria for developing an effective migration policy.

CPU and incremental memory allocation in dynamic parallelization of SQL Queries

In order to re-adjust the parallel execution of SQL queries in case of metric estimation or discretization errors, we propose an incremental parallelization method which carries out simultaneously both scheduling and mapping in co-operation with two incremental memory allocation heuristics (ParAd: parallelism degree adjustment, and MaCRelax: mapping clues relaxation) in a dynamic multi-user context. The two incremental memory allocation heuristics are integrated in the mapping method which attempt to avoid time-consuming multibucket join execution generating numerous additional I/O. A performance evaluation of the ParAd heuristic shows: (i) a significant join response time savings (from 16.11% to 35.62%), and (ii) with many complex queries, a more significant gain in response time (from 29% to 54%).

Evolution of Query Optimization Methods

Query optimization is the most critical phase in query processing. In this paper, we try to describe synthetically the evolution of query optimization methods from uniprocessor relational database systems to data Grid systems through parallel, distributed and data integration systems. We point out a set of parameters to characterize and compare query optimization methods, mainly: (i) size of the search space, (ii) type of method (static or dynamic), (iii) modification types of execution plans (re-optimization or re-scheduling), (iv) level of modification (intra-operator and/or inter-operator), (v) type of event (estimation errors, delay, user preferences), and (vi) nature of decision-making (centralized or decentralized control). The major contributions of this paper are: (i) understanding the mechanisms of query optimization methods with respect to the considered environments and their constraints (e.g. parallelism, distribution, heterogeneity, large scale, dynamicity of nodes) (ii) pointing out their main characteristics which allow comparing them, and (iii) the reasons for which proposed methods become very sophisticated.

Mobile agent cooperation methods for large scale distributed dynamic query optimization

In this paper, we study the contribution of the execution model based on mobile agents to the distributed dynamic query optimization. We propose three cooperation methods allowing to the agents, reacting to the estimation errors, to move during the runtime. The performance evaluation shows the superiority of one of these methods. It is more powerful than a standard execution starting from 30% of estimation error and has a relatively small overhead compared to a standard execution when the estimation error is small.

Location-dependent query processing under soft real-time constraints

In recent years, mobile devices and applications achieved an increasing development. In database field, this development required methods to consider new query types like location-dependent queries (i.e. the query results depend on the query issuer location). Although several researches addressed problems related to location-dependent query processing, a few works considered timing requirements that may be associated with queries (i.e., the query results must be delivered to mobile clients on time). The main objective of this paper is to propose a solution for location-dependent query processing under soft real-time constraints. Hence, we propose methods to take into account client location-dependency and to maximize the percentage of queries respecting their deadlines. We validate our proposal by implementing a prototype based on Oracle DBMS. Performance evaluation results show that the proposed solution optimizes the percentage of queries meeting their deadlines and the communication cost.

Scheduling and mapping for parallel execution of extended SQL queries

In this paper, we present an extension of PSA strategy (Parallel Scheduling Algorithm ), to determine an appropriate mapping of operations onto physical processors, taking into account the interconnection network topology of a sharednothing architecture. Performance evaluation, which relies on two benchmarks shows the efficiency of PSA strategy by comparing to Static Right-Deep strategy and to Bushy Tree Scheduling strategy. The major contributions of this work are (i) the incorporation of the mapping process into PSA strategy and (ii) the PSA strategy which provides a good trade-off between response time minimization and throughput maximization.

A mobile relational algebra

Most of the dynamic query optimization methods proposed in the literature are centralized. This centralization, in a large-scale environment, generates a bottleneck due to relatively important message exchange on a network with a weak bandwidth and strong latency. It becomes thus convenient to render autonomous and self-adaptable the query execution on a large-scale network. In this perspective, we propose a mobile relational algebra to decentralize the control of dynamic query optimization processes. Experiments, in local and large-scale distributed environments, allow: (i) to validate the proposed proactive migration policy, and (ii) to identify the efficiency intervals of proposed mobile relational algebra.

Dynamic memory allocation strategies for parallel query execution

In the decision support queries which manipulate large data volumes, it is frequent that a query constituted by several joins can not be computed completely in memory. In this paper, we propose three strategies allowing to assign the memory of a shared-nothing parallel architecture to operation clones of a query. The performance evaluation of the three strategies shows that the strategies which favor the operation clones using a lot of memory obtain a better response time than the strategy which favors the clones using little memory. The main contribution of this paper is to take into account the available memory sizes on every processor and to avoid allotting the same processor to two operation clones that must run in parallel.

Ontology-based data source localization in a structured peer-to-peer environment

In Peer-to-Peer environments, the absence of a global schema makes locating data sources a real problem. Semantic and structural heterogeneity of local schemas prevent the localization phase to find relevant data sources for a given SQL query. Useless information obtained during this phase could not only lead to incorrect answers but also it is expensive in terms of resource consumption. In this paper, we propose a method integrating domain ontology into Chord protocol. This integration provides comprehensive data exchange while carrying out efficient data source localization. Before the localization phase, the terms of the given SQL query must be written according to the domain ontology which forms the only interface to interact between peers. Chord protocol is extended by Structure Indexes that describe the relation structures. The proposed method allows extended Chord protocol to select relevant data sources and to avoid useless information. We present simulation results showing the feasibility of our method and its benefits.