HIERARCHICAL DEEP REINFORCEMENT LEARNING FRAMEWORK FOR ADAPTIVE CPU SCHEDULING IN HYBRID TRANSACTIONAL-ANALYTICAL DATABASES

Computation offloading via device-to-device communications can improve the performance of mobile edge computing by exploiting the computing resources of user devices. However, most proposed optimization-based computation offloading schemes lack self-adaptive abilities in dynamic environments due to time-varying wireless environment, continuous-discrete mixed actions, and coordination among devices. The conventional reinforcement learning based approaches are not effective for solving an optimal sequential decision problem with continuous-discrete mixed actions. In this paper, we propose a hierarchical deep reinforcement learning (HDRL) framework to solve the joint computation offloading and resource allocation problem. The proposed HDRL framework has a hierarchical actor-critic architecture with a meta critic, multiple basic critics and actors. Specifically, a combination of deep Q-network (DQN) and deep deterministic policy gradient (DDPG) is exploited to cope with the continuous-discrete mixed action spaces. Furthermore, to handle the coordination among devices, the meta critic acts as a DQN to output the joint discrete action of all devices and each basic critic acts as the critic part of DDPG to evaluate the output of the corresponding actor. Simulation results show that the proposed HDRL algorithm can significantly reduce the task computation latency compared with baseline offloading schemes.

We present a hierarchical deep reinforcement learning (DRL) framework with prominent sampling efficiency and sim-to-real transfer ability for fast and safe navigation: the low-level DRL policy enables the robot to move towards the target position and keep a safe distance to obstacles simultaneously; the high-level DRL policy is supplemented to further enhance the navigation safety. We select a waypoint located on the path from the robot to the ultimate goal as the sub-goal to reduce the state space and avoid sparse reward. Moreover, the path is generated based on either a local or a global map, which can significantly improve the sampling efficiency, safety, and generalization ability of the proposed DRL framework. Additionally, a target-directed representation for the action space can be derived based on the sub-goal to improve the motion efficiency and reduce the action space. In order to demonstrate the eminent sampling efficiency, motion performance, obstacle avoidance, and generalization ability of the proposed framework, we implement sufficient comparisons with the non-learning navigation methods and DRL-based baselines, with videos, data, code, and other supplemental material shown on our website<xref ref-type="fn" rid="fn1">¹</xref><fn id="fn1"><label>¹</label> https://github.com/zw199502/RL <inline-formula><tex-math notation="LaTeX">$\_$</tex-math></inline-formula> navigation </fn>.

Defining and structuring management views is a fundamental task of the specification of data marts and reporting systems. Today such systems are based on data warehouses. A data warehouse stores materialized views on business processes to support the management’s information requirements (cf. Inmon 1996; Inmon, Hackathorn 1994; Inmon et al. 1997). It is located on a central layer of an idealized layer oriented architecture connecting online transaction processing (OLTP) systems and components enabling online analytical processing (OLAP) (cf. Becker, Holten 1998; Chaudhuri, Dayal 1997). The latter components are intended to support navigations adequate for management users through so called multi dimensional information spaces. OLTP systems directly support the business processes and are the sources of data used by OLAP systems. Typically, the integration of OLTP systems and the data warehouse is based on tools performing so called extraction, transformation and loading tasks (ETL) (cf. Inmon 1996; Widom 1995)

In an e-commerce system, the database server performance is crucial. Dynamic caching is often used to reduce the load on the database server, which reduces the need for scalability within the server itself. A good understanding of the workload characteristics of the database server in an e-commerce environment is important to the design, tuning, and capacity planning. Using dynamic caching can dramatically alter the characteristics of requests that the database server receives, changing the dominant operations, even the workload type. The two major database workload classifications are online analytical processing and online transaction processing. Allocations for resources such as main memory can be very different, depending on the workload type. We investigate the e-commerce activity profiles resulting from dynamic caching. We find that dynamic caching does tend to alter the behavior towards transaction processing.

With the development of databases, online transaction processing (OLTP) can no longer meet the needs of end users for database query and analysis, and the simple query of large databases by SQL can not meet the requirements of end user analysis. Therefore, online analytical processing (OLAP) is proposed. concept. On the one hand, we explained the basic knowledge of OLAP, including OLAP multidimensional data concept, multidimensional data structure, multidimensional data analysis, characteristics, etc. On the other hand, we established an OLAP analysis model of the multi-temporal and spatial distribution characteristics of user energy consumption, which is refined provide support for the analysis of user energy consumption behavior.

This paper elaborates on the justification and need for integration of OLTP (online transaction processing) and OLAP (online analytical processing) data environments for the purpose of real time data analysis, and reviews potential implementation methods and architectures for integration of these disparate environments. A body of relevant literature is reviewed to answer a central question on the efficacy of OLTP and OLAP integrated environments, and the central question is made tractable by its decomposition to the essential questions that must be answered regarding the feasibility of various methods and architectures for integrating data architectures. Additional coverage is given to proposed architectures and methods for real time data warehouses in the context of OLTP and OLAP data integration. This qualitative study uses an interpretivist epistemology to reach conclusions about the literature's treatment of the topic and the justification and need for data integration to accomplish real time data analysis in the business enterprise.

A data warehouse are central repositories of integrated data from one or more disparate sources from operational data in On-Line Transaction Processing (OLTP) system to use in decision making strategy and business intelligent using On-Line Analytical Processing (OLAP) techniques. Data warehouses support OLAP applications by storing and maintaining data in multidimensional format. Multidimensional data models as an integral part of OLAP designed to solve complex query analysis in real time.

On-line transaction processing (OLTP) workloads are crucial benchmarks for the design and analysis of server processors. Typical cached configurations used by researchers to simulate OLTP workloads are orders of magnitude smaller than the fully scaled configurations used by OEM vendors to achieve world-record transaction processing throughput. The objective of this study is to discover the underlying relationships that characterize OLTP performance over a wide range of configurations. To this end, we have derived the law of database performance. Using our iron law, we show that both the average instructions executed per transaction (IPX) and the average cycles per instruction (CPI) are critical to the transaction-throughput performance. We use an extensive, empirical examination of an Oracle based commercial OLTP workload on an Intel Xeon multiprocessor system to characterize the scaling behaviour of both the IPX and the CPI. We demonstrate that across a wide range of configurations the IPX and CPI behaviour follows predictable trends, which can be accurately characterized by simple linear or piece-wise linear approximations. Based on our data, we propose a method for selecting a minimal, representative workload configuration from which behaviours of much larger OLTP configurations can be accurately extrapolated.

Hybrid storage systems are widely used in big data fields to balance system performance and cost. However, due to a poor understanding of the characteristics of database block requests, past studies in this area cannot fully utilize the performance gain from emerging storage devices. This study presents a hybrid storage database system, called HeuristicDB, which uses an emerging non-volatile memory (NVM) block device as an extension of the database buffer pool. To consider the unique performance behaviors of NVM block devices and the block-level characteristics of database requests, a set of heuristic rules that associate database (block) requests with the appropriate quality of service for the purpose of caching priority are proposed. Using online analytical processing (OLAP) and online transactional processing (OLTP) benchmarks, both trace-based examination and system implementation on MySQL are carried out to evaluate the effectiveness of the proposed design. The experimental results indicate that HeuristicDB provides up to 75% higher performance and migrates 18X fewer data between storage and the NVM block device than existing systems.

In this world, each and every activity is enriched with lot of information. Business and other organization needs information for better decision making. Business Intelligence is a set of methods, process and technologies that transform raw data into meaningful and useful information. Some of the functions of business intelligence technologies are reporting, Online Analytical Processing, Online Transaction processing, data mining, process mining, complex event processing, business performance management, benchmarking and text mining. The applications of business intelligence includes E-commerce recommender system, approval of bank loan, credit/debit card fraud detection etc., In order to obtain business intelligence from large dataset there many techniques are available in data mining such as characterization, discrimination, frequent itemset mining, outlier analysis, cluster analysis and so on. In this proposed algorithm frequent itemset mining and clustering algorithm is used to extract the information from the dataset in order to make the decision making process more efficient and to improve the business intelligence.

Ever increasing DRAM capacity has fostered the development of in-memory databases (IMDB). The massive performance improvements provided by IMDBs have enabled transactions and analytics on the same database. In other words, the integration of OLTP (on-line transactional processing) and OLAP (on-line analytical processing) systems is becoming a general trend. However, conventional DRAM-based main memory is optimized for row-oriented accesses generated by OLTP workloads in row-based databases. OLAP queries scanning on specified columns cause so-called strided accesses and result in poor memory performance. Since memory access latency dominates in IMDB processing time, it can degrade overall performance significantly. To overcome this problem, we propose a dual-addressable memory architecture based on non-volatile memory, called RC-NVM, to support both row-oriented and column-oriented accesses. We first present circuit-level analysis to prove that such a dual-addressable architecture is only practical with RC-NVM rather than DRAM technology. Then, we rethink the addressing schemes, data layouts, cache synonym, and coherence issues of RC-NVM in architectural level to make it applicable for IMDBs. Finally, we propose a group caching technique that combines the IMDB knowledge with the memory architecture to further optimize the system. Experimental results show that the memory access performance can be improved up to 14.5X with only 15% area overhead.

In the past, classical databases basically came in two flavors: optimized for write access (OLTP, online transaction processing) or read access (OLAP, online analytic processing). Typical data mining tasks, however, involve preprocessing, feature extraction, model training and cross-validation which cannot fully be categorized as either flavor. SAP’s in-memory database HANA stores all data cache-aware in main memory, allowing for rapid transactional and analytical access. While in common three-tier architectures (database, application server, client), computationally intensive applications run at the application server layer and data is loaded into the main memory of application servers, enterprise applications developed for or moved to HANA are more tightly integrated with the database. The main principle of application development for HANA is to execute data-intensive computations in the database close to the raw data in order to prevent expensive data movement. This shift in application design poses new challenges to the application developer: in order to utilize HANA efficiently, he has to think differently about how to design his application.We’ll address these challenges, discuss a real-world data analysis scenario and present some open questions in this area.

13 - Other Modeling Aspects and Trends

To catch up pace with development in processors, power consumption of a memory module is often amortized for higher bandwidth and lower latencies. As a result memory module consumes a major fraction of total power consumption in a system. This has led to low power design. This paper will evaluates three new DRAM interface, viz. MemBlaze, MemCorrect and MemDrowsy, developed at Stanford University, USA. Paper intends to extract out know-hows, tradeoffs incurred, and energy saving mode, hardware overheads and application area of the interfaces. All the three architectures work on reducing wakeup latencies of a Power-Up mode in DRAM, thus allowing the DRAM to use their Power-Down modes rigorously. The technique is capable of reducing the background power consumption up to 50%. The interface finds its application in the field which demands heavy memory usage like that of data centric applications, server memories, handling Big Data and Online Transaction Processing (OLTP) and Online Analytical Processing (OLAP).

With the ever more rapid development of information technology, computer and network infrastructures have become an indispensable part of modern life and work. In this evolving context, the importance of computer-assisted information analysis is increasingly evident, highlighting the need for a solid framework to support its further development. This paper attempts to establish a comprehensive, multi-dimensional technological framework to improve capacity in this crucial area. The proposed framework covers core technology areas such as databases, data warehouses, online transaction processing (OLTP), business intelligence (BI), online analytical processing (OLAP) and decision support systems (ss). By combining these fundamentals with cutting-edge technologies, including data extraction, social network analysis, and massive data analysis and advanced online analytical processing, the framework aims to create a holistic system for computer-aided information analysis. The system aims not only to optimize the efficiency and effectiveness of data processing and analysis, but also to promote innovation in the way information is used for decision-making and strategic planning, contributing to the continued development and impact of all industries.