Database Systems Journal

ISSN 2069 - 3230

The journal is published under the sponsorship of
The Bucharest University of Economic Studies
and it is produced by the university's own publishing division,
The Bucharest University of Economic Studies Publishing House

Database Systems Journal, Vol. VI, Issue 4/2015
Issue Topic: Big Data Analytics

Open PDF Journal


1. NoSQL Key-Value DBs Riak and Redis (p. 3-10)
Cristian Andrei BARON, University of Economic Studies, Bucharest, Romania
In the context of today's business needs we must focus on the NoSQL databases because they are the only alternative to the RDBMS that can resolve the modern problems related to storing different data structures, processing continue flows of data and fault tolerance. The object of the paper is to explain the NoSQL databases, the needs behind their appearance, the different types of NoSQL databases that current exist and to focus on two key-value databases, Riak and Redis.
Keywords: NoSQL Databases, Key-Value, Riak, Redis, RDBMS.
2. Boarding to Big data (p. 11-17)
Oana Claudia BRATOSIN, University of Economic Studies, Bucharest, Romania
Today Big data is an emerging topic, as the quantity of the information grows exponentially, laying the foundation for its main challenge, the value of the information. The information value is not only defined by the value extraction from huge data sets, as fast and optimal as possible, but also by the value extraction from uncertain and inaccurate data, in an innovative manner using Big data analytics. At this point, the main challenge of the businesses that use Big data tools is to clearly define the scope and the necessary output of the business so that the real value can be gained.
This article aims to explain the Big data concept, its various classifications criteria, architecture, as well as the impact in the world wide processes.
Keywords: Big data, Predictive Analytics, Data mining, Internet of Things.
3. The Importance of Data Warehouses in the Development of Computerized Decision Support Solutions. A Comparison between Data Warehouses and Data Marts (p. 18-26)
Alexandru Adrian ŢOLE, Romanian - American University, Bucharest, Romania
In the last decade, the amount of data that an organization processes and stores has grown exponentially. In most cases, the data stored is used to support the business process through accurate and up-to-date information about the business environment and activity of the company. In order for a company's managers to be capable of generating the reports they need to make decisions, one needs a computer system able to store complex and very large quantities of data. At the same time, for the development of such an information system, one must take into account the cost of it.
Keywords: Data Warehouse, Data Mart, Top-down, Bottom-up, database, architecture, management system.
4. Optimizing memory use in Java applications, garbage collectors (p. 27-32)
Ştefan PREDA, Oracle, Bucharest, Romania
Java applications are diverse, depending by use case, exist application that use small amount of memory till application that use huge amount, tens or hundreds of gigabits. Java Virtual Machine is designed to automatically manage memory for applications. Even in this case due diversity of hardware, software that coexist on the same system and applications itself, these automatic decision need to be accompanied by developer or system administrator to triage optimal memory use. After developer big role to write optimum code from memory allocation perspective , optimizing memory use at Java Virtual Machine and application level become in last year's one of the most important task. This is explained in special due increased demand in applications scalability.
Keywords: Java Virtual Machine, garbage collector, Concurrent Mark Sweep (CMS), G1 GC, Shenandoah an Ultra-Low-Pause-Time Garbage Collector.
5. Data mining in healthcare: decision making and precision (p. 33-40)
Ionuţ ŢĂRANU, University of Economic Studies, Bucharest, Romania
The trend of application of data mining in healthcare today is increased because the health sector is rich with information and data mining has become a necessity. Healthcare organizations generate and collect large volumes of information to a daily basis. Use of information technology enables automation of data mining and knowledge that help bring some interesting patterns which means eliminating manual tasks and easy data extraction directly from electronic records, electronic transfer system that will secure medical records, save lives and reduce the cost of medical services as well as enabling early detection of infectious diseases on the basis of advanced data collection. Data mining can enable healthcare organizations to anticipate trends in the patient's medical condition and behaviour proved by analysis of prospects different and by making connections between seemingly unrelated information. The raw data from healthcare organizations are voluminous and heterogeneous. It needs to be collected and stored in organized form and their integration allows the formation unite medical information system. Data mining in health offers unlimited possibilities for analyzing different data models less visible or hidden to common analysis techniques. These patterns can be used by healthcare practitioners to make forecasts, put diagnoses, and set treatments for patients in healthcare organizations.
Keywords: Data Mining, Big Data, Knowledge Discovery.