Workload adaptation in autonomic DBMSs

Modern network computing is in the process of changing from manual management to autonomic computing. As the core of network computing, database management system with autonomic computing must be able to participate in the process to meet the performance requirements of the network applications. This paper studies on workload adaptation technology. In the workload adaptation framework, the performance forecasting component is the base of developing workload control scheme. Layered queuing network modeling techniques was used to establish the performance model of DBMS. A test platform based on this architecture has been built and been tested. Test result shows that this framework has played an effective role in workload control, and improved client satisfaction rate and the application performance of the network.

In today's data server environments, multiple types of workloads can be present in a system simultaneously. Workloads may have different levels of business importance and unique performance goals. An autonomic workload management system controls the flow of the workloads to help the database management system (DBMS) meet the performance goals. A task of the autonomic workload management system is to prevent congestion in the DBMS, which can result in severe degradation in overall system performance. Autonomic workload management should detect that a system is becoming congested and then act to restore normal system operation. In this paper, we describe an approach to identify a set of database monitor metrics that can serve as indicators for potential congestion in a specific scenario. We present experiments to illustrate two cases of congestion in a DB2® DBMS and use our approach to derive the indicators.

Service providers want to reduce datacenter costs by consolidating workloads onto fewer servers. At the same time, customers have performance goals, such as meeting tail latency Service Level Objectives (SLOs). Consolidating workloads while meeting tail latency goals is challenging, especially since workloads in production environments are often bursty. To limit the congestion when consolidating workloads, customers and service providers often agree upon rate limits. Ideally, rate limits are chosen to maximize the number of workloads that can be co-located while meeting each workload's SLO. In reality, neither the service provider nor customer knows how to choose rate limits. Customers end up selecting rate limits on their own in some ad hoc fashion, and service providers are left to optimize given the chosen rate limits. This paper describes WorkloadCompactor, a new system that uses workload traces to automatically choose rate limits simultaneously with selecting onto which server to place workloads. Our system meets customer tail latency SLOs while minimizing datacenter resource costs. Our experiments show that by optimizing the choice of rate limits, WorkloadCompactor reduces the number of required servers by 30--60% as compared to state-of-the-art approaches.

The increasing complexity of database management systems (DBMSs) and their workloads means that manually managing their performance has become a difficult and time-consuming task. Autonomic computing systems have emerged as a promising approach to dealing with this complexity. Current DBMSs have begun to move in the direction of autonomic computing with the introduction of parameters that can be dynamically adjusted. A logical next step is the introduction of self-tuning technology to diagnose performance problems and to select the dynamic parameters that must be adjusted. We introduce a method for automatically diagnosing performance problems in DBMSs and then describe how this method can be incorporated into current DBMSs using the concept of reflection. We demonstrate the feasibility of our approach with a proof-of-concept implementation for DB2 universal database.

Increasing workload on multi-user database management system (DBMS) can cause maximum system's throughput. With further increase of workload, throughput dramatically decreases and system performance fall down; then system starts to thrash. Existence of an admission control component for managing user's requests is necessary. Admission control component in DBMS has two main tasks. These tasks are Prevention of increasing user's communication requests to system more than a threshold and prevention, deferment or execution of received requests after optimization step and determination of request requirements. The problem is determination of allowable number of communications to DBMS. Can we predict and determine conditions to providence time and resource before reach to optimization step? In this paper we use data mining techniques to predict and describe admission control policies. According to this purpose we recommend an Admission Logging system to log existing actions before end of query processor procedures. We describe situation, structure, and changes that we need in system. Then we describe the circumstance of knowledge extraction on Admission Log.

Virtualization provides increased control and flexibility on how resources are allocated to applications. However, common resource provisioning mechanisms do not fully use these advantages; either they provide limited support for applications demanding quality of service, or the resource allocation complexity is high. To address these issues we developed Themis, a market-based application management platform. By limiting the coupling between the applications and resource management, Themis can support diverse types of applications and performance goals while ensuring maximized resource usage. In this paper we present the performance of Themis when users execute batch applications with different Service Level Objectives such as deadlines.

A multidatabase system (MDBS) is a facility that allows users access to data located in multiple autonomous database management systems (DBMSs). In such a system, global transactions are executed under the control of the MDBS. Independently, local transactions are executed under the control of the local DBMSs. Each local DBMS integrated by the MDBS may employ a different transaction management scheme. In addition, each local DBMS has complete control over all transactions (global and local) executing at its site, including the ability to abort at any point any of the transactions executing at its site. Typically, no design or internal DBMS structure changes are allowed in order to accommodate the MDBS. Furthermore, the local DBMSs may not be aware of each other and, as a consequence, cannot coordinate their actions. Thus, traditional techniques for ensuring transaction atomicity and consistency in homogeneous distributed database systems may not be appropriate for an MDBS environment. The objective of this article is to provide a brief review of the most current work in the area of multidatabase transaction management. We first define the problem and argue that the multidatabase research will become increasingly important in the coming years. We then outline basic research issues in multidatabase transaction management and review recent results in the area. We conclude with a discussion of open problems and practical implications of this research.

Capacity planning for IaaS cloud providers offering multiple service classes

Functionality, complexity, heterogeneity and dynamism in computing environment are increasing day by day. This enhanced utility of computers has a profound impact on the systempsilas brittleness, manageability and security. Self-management is important in systems, networks, communication as well as in Database Management Systems (DBMSs). Autonomic computing reduces the problems and increases accuracy and efficiency of the DBMSs. Recent years have seen an upsurge in research related to incorporation of autonomic computing within the computing systems. IBM is working on autonomic computing in systems and DBMSs from last many years. The paper discussed the autonomic features and tools in DB2 from the literature available up to 2002. Our paper presents an extension of that paper which covers available autonomic features, components, utilities and tools up till now. A comparison of DB2 with Oracle is presented w.r.t autonomic computing. We have also identified the optional and essential human intervention (HI) in autonomic components, utilities and tools of IBM Universal Database DB2 that reveals the degree of autonomic computing in these.

System-of-Systems (SoS) offer unprecedented potential for new types of emerging services, which significantly exceed the capabilities of the constituting systems. SoS in safety-critical domains (e.g., medical applications, smart grid, disaster recovery, defense) are prominent examples, but they have stringent real-time and reliability requirements. Therefore, a suitable temporal and spatial allocation of resources is required both within each constituent system and in the wide area networks between them. This paper introduces an algorithm for admission control and resources’ allocation, which considers these requirements and the autonomy of the constituent systems. To simulate a realistic admission control and resources’ allocation process of a typical SoS network, a simulated case study with eight constituent systems, six services, and twenty-five processes/requests is developed. The suggested admission control and resources’ allocation process’s performance is measured in terms of gain in the execution time and blockage probability. A sensitivity analysis is carried out to evaluate the influence of the number of constituent systems and the number of services sought by the received processes/requests on the efficacy of the proposed process. The results show that the proposed admission control and resources’ allocation process have very low blockage probability, high gain in the execution time, and high resources’ utilization.

The rapid growth of enterprise data and the increasing complexity of modern database systems have necessitated a shift from traditional manual database management to autonomous, AI-driven solutions. AI-driven autonomous database management systems (ADBMS) leverage machine learning, predictive analytics, and automation to optimize database performance, reduce administrative overhead, and enhance scalability in enterprise IT environments. Traditional database management approaches often suffer from inefficiencies related to query performance, indexing, workload tuning, and anomaly detection, leading to increased operational costs and performance bottlenecks. This paper explores the key components of AI-driven autonomous database management, focusing on self-tuning mechanisms, predictive query optimization, and intelligent indexing techniques. Self-tuning capabilities leverage AI to analyze workloads, optimize resource allocation, and dynamically adjust system parameters to maintain peak efficiency. Predictive query optimization utilizes deep learning algorithms to enhance query execution plans, reduce latency, and anticipate performance issues before they impact business operations. Additionally, intelligent indexing applies machine learning techniques to automate index selection, adaptation, and maintenance, ensuring optimal data retrieval and reducing query processing times. By integrating these AI-driven mechanisms, enterprises can achieve greater operational efficiency, improved database reliability, and reduced human intervention in performance tuning. The study also addresses security, compliance, and reliability concerns associated with autonomous database management, proposing best practices for AI-driven data governance. Future research directions include the integration of quantum computing for database acceleration, AI-driven anomaly detection for enhanced cybersecurity, and the application of reinforcement learning for real-time database optimization. This paper provides a strategic roadmap for enterprises looking to adopt AI-driven autonomous database solutions to drive innovation and competitive advantage.

The article contains an analytical review of developments in database technologies, made on the basis of reports prepared by the results of eight meetings of database specialists held throughout 1988–2013. Objects of the analysis are most interesting predictions given in the reports: their realism, accuracy, pragmatism or, vice versa, utopianism or opportunism. The article consists of two parts. Part 1 is devoted to analysis and evaluation of predictions made in the reports of the four earlier meetings held in 1988, 1990, 1995, and 1996. These predictions are about creation, development and uses of decision support systems, database appliances, graphic processing units, operating systems, interface for structured query language, database applications, information distribution, universal database management systems, query optimization criteria, intellectual analysis of database within database management systems. A detailed description of research themes in the field of databases, which got the priority status in that time, is given: recording and computation of data, security and confidentiality of data, replication and harmonization of data, structuring of data, intellectual analysis of data, data warehouses. Part 2 is devoted to an analytical review of the predictions contained in the reports on the meetings held in 1998, 2003, 2008, and 2013. The predictions are about self-adjustment of database systems, rethinking of the traditional database architecture as a result of new hardware capabilities. They make special emphasis on the feasibility of manipulations with structured and unstructured data within DSS architecture, support of Big Data technology, with outlining the themes of research aimed at implementation of its potential.

Moving object database is a collection of geometric data that changing over time. The existence of a moving object database allows database users to express a series of predicates that state events or changes that happened during the movement of a moving object. The class of queries that involve a series of predicates is called the Spatio-temporal pattern (STP) queries. STP query has temporal order constraints on the fulfillment of predicates involved. The implementation of the STP query is unavailable in existing database management systems (DBMSs). Developing a tool to express the STP query needs to be done. This is to allow any kind of questions that could not be asked before.STP query is expressed using the STP predicate which has a design consisting of two parts. The first defines lifted predicates that accept moving object arguments and report the time intervals, during which they are fulfilled. The second part is to define temporal constraints on the predicate fulfillments. Temporal constraints are defined as temporal relationships between pairs of time intervals.Implementation of STP queries is done by providing data structures for moving objects, methods for evaluating operations on moving objects, and functions for defining STP predicates. The implementation is collected into a single package called extension. Extension is a form of tool built on the DBMS to handles STP queries in a moving object database. The extension is built on PostgreSQL which provides spatial data types and operations through PostGIS. The testing of the tool is done by executing several STP queries. The test results show that the tool built successfully handles STP queries in a moving object database.

Wireless networks are playing an increasingly important role for global communications. Many resource allocation mechanisms have been proposed to efficiently utilize the limited radio resources in wireless networks to support a large number of mobile users with a diversity of applications. Among them, pricing frameworks that provide incentives to users to maximize their individual utility while optimizing allocation of network resources have attracted a lot of attention recently. Nevertheless, most of these pricing schemes require dynamic charging rates and may be too complex for wide acceptance by users, as most users would prefer relatively simple charging schemes. Moreover, use of a pricing framework to facilitate resource planning and future expansion at the service provider's side has not yet been widely considered. In this paper, we propose Integrated Multiple Time Scale Control (IMTSC), a novel incentive engineering mechanism to facilitate resource allocation and network planning. Over different time scales, IMTSC combines the functions of network capacity planning, admission control for resource allocation, and tracking of users' instantaneous traffic demands. The proposed mechanism is applied for access control at a congested access point in a wireless network. By decomposing the original problem into distributed optimization problems that are solved locally by the service provider through adjusting charging rate and remotely by individual users by appropriately changing her service requests, we show that maximization of user's utility and increase of network efficiency can be simultaneously achieved. Results from extensive simulations demonstrate the effectiveness of the proposed IMTSC mechanism.

In cloud computing model, resource sharing introduces major benefits for improving resource utilization and total cost of ownership, but it can create technical challenges on the running performance. In practice, orchestrators are required to allocate sufficient physical resources to each Virtual Machine (VM) to meet a set of predefined performance goals. To ensure a specific service level objective, the orchestrator needs to be equipped with a dynamic tool for assigning computing resources to each VM, based on the run-time state of the target environment. To this end, we present LOOPS, a multi-loop control approach, to allocate resources to VMs based on the service level agreement (SLA) requirements and the run-time conditions. LOOPS is mainly composed of one essential unit to monitor VMs, and three control levels to allocate resources to VMs based on requests from the essential node. A tailor-made controller is proposed with each level to regulate contention among collocated VMs, to reallocate resources if required, and to migrate VMs from one host to another. The three levels work together to meet the required SLA. The experimental results have shown that the proposed approach can meet applications' performance goals by assigning the resources required by cloud-based applications.