Rivero L.Encyclopedia of database technologies and applications.2006

and partitioning depends on the application, but it is possible to determine general data access patterns in case of WfMS (Schuster & Heinl, 1997).

Metadata Description and Database

Schemas

Metadata descriptions are employed by database systems to make them aware of other database needs (Bernstein et al., 1998) and also to locate relevant data in distributed systems, providing enough background information to judge the suitability and trustworthiness of the data sources. It is an important technology concerning collaboration, flexibility, and evolution in workflow systems.

Database schemas reflect the rules governing processes and their data. Schema evolution (Lerner, 2000) gives the user the ability to dynamically change the schema of a data source. This is important to preserve flexibility and guarantee consistency in changing environments.

Transaction Models

New transaction models, allowing flexible ways of dealing with consistency and recovery is paramount to support collaborative work. Nested transactions and open nested transactions are well-known advanced transaction models, allowing composite units of work to be reflected as transaction structures. A survey of advanced transactions models can be found in Barghouti and Kaiser (1991). Some workflow applications may require the definition of application-specific transaction models (Georgakopoulos, Hornick & Manola, 1996). Multiuser transaction for cooperative applications is described in Wieczerzycki (1999).

Database Synchronization Techniques

Concurrency requires synchronization to guarantee the integrity of concurrent data and processes. A number of synchronization techniques, such as fetch-and-add, set- and-test, and wait-free synchronization, are already being used by database systems (Herlihy, 1991) and may be applied by WfMS.

FUTURE TRENDS

There should be advances in all areas related to the flexible, cooperative, and diversified ways of doing business. This points to an increasing use of multiple WfMS engines to support distributed workflow process execution. We should expect more research in the areas of flexible coordination, awareness provision, team scalability, and integration of workflow systems with

computer support for cooperative work (CSCW) and content management technologies (Geogakopoulos, 2004).

The overall trend is to increase flexibility and to incorporate controlled inconsistency, allowing it to exist for a while and be solved later in an application by application basis (Dourish, 1998). For example, we can have synchronization methods designed to minimize divergence between source data and cached copies, instead of trying to maintain exact synchronization (Olston & Widom, 2002). Customized transaction models (Geogakopoulos, 2004), advances on metadata description (Roantree, 2002), and schema evolution (Pittas, Jones & Gray, 2001) should all have a positive impact on collaborative work.

CONCLUSION

A WfMS implemented on top of a database system is a special database application. Therefore, it helps to understand how advances on databases as a supporting technology may be applied to build more useful workflow applications and, on the other hand, how workflow application needs may drive the improvement of database technologies.

Nevertheless, workflow applications are highly hu- man-centered and should not be limited by technology. Organizational requirements have to be taken into consideration when developing workflow applications and in choosing an appropriate WfMS (Silva & Pinheiro, 2003). These requirements may impose restrictions that will never be completely solved by any supporting technology. For example, the most challenging kind of diversity is cultural diversity. It may be ameliorated using techniques to deal with semantic heterogeneity, but there is more to culture than differences in the interpretation of terms.

REFERENCES

Barghouti, N., & Kaiser, G. (1991). Concurrency control in advanced database applications. ACM Computing Surveys, 23(3), 269-317.

Bernstein, P., Brodie, M., Ceri, S., DeWitt, D., Franklin, M., Garcia-Molina, H., et al. (1998). The asilomar report on database research. SIGMOD Record, 27(4), 7480.

Dourish, P. (1998). Using metalevel techniques in a flexible toolkit for CSCW applications. ACM Transactions on Computer-Human Interaction, 5(2), 109-155.

Database Support for Workflow Management Systems

Georgakopoulos, D. (2004). Teamware: An evaluation of key technologies and open problems. Distributed and Parallel Databases, 15(1), 9-44.

Georgakopoulos, D., Hornick, M., & Manola, F. (1996). Customizing transaction models and mechanisms in a programmable environment supporting reliable workflow automation. IEEE Transactions on Data and Knowledge Engineering, 8(4), 630-649.

Herlihy, M.P. (1991).Wait-free synchronization. ACM Transactions on Programming Languages and Systems, 13(1), 124-149.

Lerner, B.S. (2000). A model for compound type changes encountered in schema evolution. ACM Transactions on Database Systems, 25(1), 83-127.

Olston, C., & Widom, J. (2002). Best-effort cache synchronization with source cooperation. Proceedings ACM International Conference on Management of Data (pp. 7384).

Pittas, N., Jones, A.C., & Gray, W.A. (2001). Evolution support in large-scale interoperable systems: A metadata driven approach. Proceedings of the 12th Australasian Conference on Database Technologies (pp. 161-168).

Roantree, M. (2002). Metadata management in federated multimedia systems. Proceedings of the 13th Australasian Conference on Database Technologies (Vol. 5, pp. 147155).

Schuster, H., & Heinl, P. (1997). A workflow data distribution strategy for scalable workflow management systems.

ACM Symposium on Applied Computing, 174-176.

Silva, L.P., & Pinheiro, F.A.C. (2003). Eliciting requirements for identifying workflow categories. Proceedings of the VI Workshop on Requirements Engineering (WER’03) (pp. 16-31).

Stohr, E.A., & Zhao, J.L. (2001). Workflow automation: Overview and research issues. Information Systems Frontiers, 3(3), 281-296.

van der Aalst, W., ter Hofstede, A.H.M., Kiepuszewski, B., & Barros, A.P. (2003). Workflow patterns. Distributed and Parallel Databases, 14, 5-51.

WfMC. (2004). Workflow management coalition. Retrieved April 27, 2005, from http://www.wfmc.org

Wieczerzycki, W. (1999). Database model for Webbased cooperative applications. Proceedings of the 8th Conference on Information and Knowledge Management (pp. 131-138).

KEY TERMS

D

The definitions below are based on Georgakopoulos (2004), Schuster and Heinl (1997), Stohr and Zhao (2001), and WfMC (2004).

Content Management Systems: Provide tools for organizing, delivering, and sharing documents and images. Usually used in conjunction with CSCW systems or workflow systems.

CSCW: Computer-Supported Cooperative Work. Provides tools for supporting people working together, for example, video and audio conferences, group calendar, e- mail, and text chat. Differs from workflow applications in having more flexibility and less coordination.

Data Partitioning: A storage technique through which each data item is assigned to exactly one node. Data operations accessing different data partitions can be executed in parallel. However, if one operation needs to access more than one data partition, the execution is more complicated.

Data Replication: A storage technique through which some nodes have copies of the same data. Replication of data is a common method to improve read performance but is rather problematic if data is often updated.

Metadata Description: Abstract data description used to describe concrete data items. Usually contains descriptions of data structures, relationships to other data items, location, scope, and other features of the data item.

Process Enactment: A WfMS is said to enact the real-world process for each process instance.

Task Assignment: The assignment of a task to an agent responsible for it, made by the WfMS. The agent may be a person or a process. When it is a person, the assignment usually implies a notification to the agent; when it is a process, the assignment usually results in its execution.

Workflow Instance: Also, process instance. Each execution instance of a process managed by the WfMS.

Indranil Bose

University of Hong Kong, Hong Kong

Wang Ping

University of Hong Kong, Hong Kong

Mok Wai Shan

University of Hong Kong, Hong Kong

Wong Ka Shing

University of Hong Kong, Hong Kong

Yip Yee Shing

University of Hong Kong, Hong Kong

Chan Lit Tin

University of Hong Kong, Hong Kong

Shiu Ka Wai

University of Hong Kong, Hong Kong

INTRODUCTION

Owing to the rapid development of mobile technology over the past few decades, there have been many different kinds of mobile devices emerging in the market, and most of them work with databases seamlessly. Mobile phone gaming, downloading of ringtones, and e-calen- daring are some of the prominent examples of mobile applications that require the close integration of mobile devices with databases. Mobile devices take various forms and configurations. The packaging, form factors, hardware platforms, operating system support, and functional capabilities vary across these devices. There are, however, many common attributes shared by the devices, such as notebook computers, pen-based computers, handheld computers, and the like, all of which are used in mobile computing. These devices can be categorized into the following categories according to their functionalities and features, as detailed in Dhawan (1997). They are:

•notebook computers

•personal digital assistants

•tablet computers

•hybrid mobile devices

•mobile phones

In this article, we focus on personal digital assistants (PDA) and mobile phones as they are the most popular and commonly used mobile devices in the industry.

Mobile Computing Applications

Mobile applications include basic applications like datebook, address book, to-do list, and memos and also horizontal and vertical industry applications that mainly fall into the following three categories (Dhawan, 1997):

•Shrink-wrapped horizontal industry mobile computing applications that can be used in broad segments of various industries, e.g., electronic mail, electronic messaging via paging, and sales force automation.

•Generic horizontal industry applications requiring extensive customization, and these include database access from an information server, com- puter-aided dispatch (CAD), and intrasite and intersite mobility applications among others.

•Vertical industry applications include the applications that are specific to industries like insurance, banking, airlines, government, utilities, and transportation, e.g., finance industry insurance and financial planning, and stock trading.

Databases for Mobile Applications

The diverse variety of the types of mobile applications demonstrates the reach of mobile computing into almost every facet of personal and business life. One of the applications that is gaining popularity is mobile e-com- merce. Mobile e-commerce refers to commercial activities performed electronically. An example of this is an online shopping mall (via the mobile devices to the Internet). Mobile commerce is one of the most popular applications these days in addition to obtaining stock quotes, directions, weather forecasts, and airline flight schedules from mobile devices (Munusamy & Hiew, 2004).

Comparison Between Mobile Devices

and Desktop Computers

Compared to desktop computers, mobile devices have small memory, low computing capabilities, limited interaction facilities, and limited display and network processing capabilities. With recent technological advancements, hybrid devices combining the functionality of mobile phones together with PDAs have been developed. The differences are mainly attributed to their hardware design and system configurations. Table 1 compares desktop computers, PDAs, and mobile phones with respect to their processing power, memory, storage capacity, connection speed, and display. The data presented is current as of June 22, 2004. Data related to specifications for the desktop, PDA, and mobile phone have been downloaded from the Web sites of Dell Inc. and Nokia Inc. and relate to the Dell Dimension 8400 Desktop, the Dell Axim X3 Pocket PC 400 MHz WiFi, and the Nokia 7610, respectively.

From Table 1, it can be observed that mobile devices have smaller memory size and storage capacity as well as display size than desktop computers. So, the amount of data that can be transferred and displayed at a time is less than that of desktop computers. Furthermore, the

Table 1. Comparison between a desktop computer and mobile devices

processing power of mobile devices is usually limited

when compared with desktop computers. The amount of D data that can be processed at a given time is also small.

Also, mobile devices have lower connection speeds and less stable network connections. They must have ways to overcome these deficiencies in order to ensure good performance in retrieving data from remote databases.

Challenges for Mobile Devices

Some of the challenges faced by mobile devices when connecting to remote databases include challenges in network connectivity, data transmission, security, and data consistency.

•Network connectivity: Mobile devices usually work in an unstable network environment. The network stability is affected by many factors, such as weak signal and strong interference. Without physical network connections, mobile devices often lose connection with the network.

•Data transmission: Wireless networks have limited bandwidth compared to traditional cable networks. The slow transmission speed imposes problems in uploading and downloading data. Large network latency constraints also result in long response time.

•Security: Any message between the database system and mobile devices is sent over the air, and it is possible for hackers to sniff the message and perform eavesdropping. Advanced encryption and user authentication technology is needed to prevent any such types of hacking activities.

•Data consistency: Database applications apply extensive caching and replication to boost performance, which can lead to possible data inconsistency. Mobile devices with little memory storage and slow connection speed cannot obtain all the information from the central database system instantly. The narrow bandwidth of the devices also affects immediate updates from the mobile devices to the database server. It is thus quite difficult to keep data consistent between mobile devices and the database server.

The objective of this article is to give a brief overview on the design of databases for mobile applications and to describe how the database design is currently being done for a successful mobile application called mBroker that is operational in Hong Kong. The article provides a description of the functionalities of the mBroker system and highlights the database design being used by the mBroker solution at the present time.

Figure 1. A typical database architecture for mobile applications

Databases for Mobile Applications

Everyplace and Oracle Lite (Viellard, 2001) are examples of micro database engines.

BACKGROUND

The database design for mobile applications is different from normal database applications running on personal computers. Due to the limited hardware configurations and network settings of mobile devices, database vendors usually provide special database systems and APIs (application programming interfaces) tailored for mobile devices.

Database manufacturers like Oracle, IBM, and Sybase usually follow a similar architecture to build mobile database applications. The main components include a micro database engine, synchronization middleware, and wireless networking. The difference is often the naming of different components. Figure 1 depicts the typical database architecture for mobile applications.

Micro Database Engine

Mobile devices have limited hardware in terms of processing power, memory size, and battery life. A normal database engine requires a minimum 20 MB memory to operate, which is unavailable for mobile devices. Therefore, database vendors develop robust micro database engines for most mobile devices.

Micro database engines only focus on the most frequently used functionalities which are relevant to mobile applications. These include basic SQL statements, Join, Group By, Order By, scrollable cursors, and simple primary key and foreign key operations. For high-end mobile devices, advanced indexing features are sometimes included to improve performance. Database operations which are rare and less useful are removed from the micro engine. For example, view creation, subqueries, stored procedures, triggers, and user-defined functions are not provided in micro database engines. The DB2

Synchronization Middleware

The network capabilities of mobile devices are limited. Real-time online database access requires a large and stable network bandwidth, which is expensive and not always required for mobile applications. Thus, database vendors employ extensive caching and synchronization techniques in mobile database applications. Caching is to preload data to the devices for offline browsing. Users can then browse data seamlessly even if the wireless network is unstable or even unavailable. If instant update to the central server is not necessary, data modification operations are further optimized. Updates are made to the data on the devices first, and the modifications are not reflected on the server database immediately. Data are updated to the remote central database later by synchronization.

Synchronization is done by middleware which is sometimes called a synchronization server or mobile server. Advanced synchronization methods keep data in both mobile devices and the central database consistent. The middleware acts as a middleman between the mobile devices and the central database. It collects changes in the mobile devices and executes appropriate SQL statements to update the central database. At the same time, the middleware also propagates updates in the central database to the mobile devices.

Wireless Networking

Different types of mobile devices communicate using different network protocols. Database vendors usually include a number of network protocol supports in their products, for example, GSM, IEEE 802.11, and Bluetooth.

Security

To overcome the security problems of mobile database applications such as eavesdropping, the mobile database architecture needs to support both username/password authentication and encrypted communication based on the Secure Sockets Layer (SSL) protocol and other popular encryption algorithms.

In spite of the well designed architectural model, there still exist a number of limitations in the integration of the databases with mobile devices.

Databases for Mobile Applications

Literature Review

Previous research related to database design for mobile applications has involved various mechanisms for avoiding compromise in the performance of the database due to the use of the wireless network. Some of the techniques used involve reducing the number of data exchanged over the wireless network and providing a data cache on the mobile host (Chan, Si, & Leong, 1998). To address the challenges related to maintenance of data consistency for mobile data access, several techniques have been suggested in the literature. These have ranged from transaction management (Mazumdar & Chrysanthis, 1999) and concurrency control for mobile databases (Prabhu, Ray, & Yang, 2004) to replication of mobile databases which are allocated on the fixed network (Budiarto, Harumoto, Tsukamoto, Nishio, & Takine, 1998). It is argued in Budiarto, Nishio, and Tsukamoto (2002) that without replication, mobile databases have a very low availability, and it is shown using simulation that the performance of replication strategies depends on various factors such as network size, mobility, access ratio, and access concentration. Another important issue for mobile databases is how to provide consistent results for location-specific continuous queries, the likes of which may be encountered when navigating road maps using mobile devices. In Gok and Ulusoy (2000), several approaches are compared in terms of relative performance for providing answers to location-dependent queries from mobile users. An analytical model based on the idle replacement policy is described by Hung, Lin, Peng, and Yang (2001) to solve the problem of overflows of visitor location registers for mobile databases. For a detailed discussion on the various issues related to database design for mobile databases with respect to factors such as mobile location data management, transaction processing and broadcast, cache management and replication, query processing, and mobile Web services, the interested reader may refer to Barbara (1999), Madria, Mohania, Bhowmick, and Bhargava (2002), and Yang, Bouguettaya, Medjahed, Long, and He (2003).

MAIN THRUST

Sixteen Hong Kong brokerage firms are currently improving their productivity and customer satisfaction levels by using the “mBroker” solution offered jointly by Heracle Technologies Limited and Hutchison Telecom, built on the Palm OS platform for Palm handhelds (Lai, Tam, & Lemaitre, 2004). The mBroker solution provides a secure trading platform to remotely access stock

Problem Description

It is time-consuming and labor-intensive for investors to rely on desktop computers or consult brokerage agents to obtain the latest stock quotes or the trading history of any particular stock index, as well as to place an order. Mistakes such as overlooked orders, data entry errors, or delays due to congested telephone lines often occur when dealing with an agent.

It is critical to guarantee high speed and accuracy for this type of situation. Speeding up the order processing and accessing real-time information without compromising accuracy are goals that all brokers strive for to retain their competitive edge.

The mBroker Solution

mBroker—an innovative, secure, and cost-effective wireless stock trading system with specific design for PDA stock trading—allows PDA users to get stock quotes and trade stocks at any time in any place in the world. More importantly, it is totally secure as it uses the “Hongkong Post Mobile e-Cert” and also Oracle8i. With mBroker, investors can place an order remotely via an intuitive touch screen interface of Palm handhelds, without the need to contact the agent.

Functionalities of mBroker

There are a diverse range of services related to stock trading activities provided by mBroker on Palm. These include:

•Place, modify, and cancel orders.

•Access real-time stock quotes.

•Keep track of the Hang Seng Index.

•View order status and transaction history.

The mBroker application can be run on Palm OS 3.5 or above and only occupies approximately 190 Kb of memory space. What makes this solution attractive is the minimal deployment cost. Since there is no special software or hardware requirements for participating brokerage firms, their customers can enjoy the service simply by subscribing to the mBroker service. The user interface is very user-friendly since it has a similar look and feel to any other Palm application. Both English and Chinese versions of this software are available to the users.

Figure 2. Application interface of mBroker on a PDA device

Architecture of mBroker

Figure 3 shows a simplified view of the mBroker application and the database system architecture. The Oracle database stores stock information and statistics. The stock data includes stock codes, stock names, and stock prices. In order to help the investors in making investment decisions, the database also stores information such as past selling prices of the stock, the high/low price information, and transaction volume statistics. A subset of this information is loaded onto the Palm for fast offline browsing.

Transaction details are also required for order tracking and enquiry. Therefore, all the transaction details are logged in the Palm and the central database. These include the broker ID, status, stock type, price, quantity, transaction timestamp, etc.

The mBroker application follows an architecture that is similar to that discussed earlier. The stock and

Figure 3. A simplified view of the architecture of mBroker

mBroker

Databases for Mobile Applications

transaction information is originally stored in a central Oracle 8i database server. When a mobile device with mBroker wants to retrieve stock quotes from the central database server, it will send requests to a middleware named Oracle Lite Mobile Service in order to subscribe to the information (Viellard, 2001). This middleware is installed in a server machine and acts as a middleman between different mobile devices and the central database server. It obtains the required information from the database server and sends it back to the mobile devices. The middleware is responsible for synchronization, keeping the mobile devices informed of any changes on the database server. When the user buys or sells stock via mBroker, requests are sent from the mobile devices to the middleware. The middleware then updates the central database accordingly. The middleware is capable of receiving thousands of requests at the same time and decides the sequence of processing the transaction requests following predefined business rules written in PL/SQL.

Choice of Database

Heracle Technologies Limited chose to use Oracle8i as the central component because of mobile service provided by Oracle8i Lite. Oracle8i Lite is accessed by users through the application server and provides the necessary workspace for end users to request context switching between online and offline modes. It also automatically initiates the necessary two-way replication of data and applications between server and client, depending on changes of mode. When it comes to stock order transactions, it is imperative to ensure a maximum level of security. Not all proprietary databases, e.g., DB2, can support a public key infrastructure (PKI; Browder, 2002); Oracle8i supports PKI and therefore this was the obvious choice for mBroker.

Security of mBroker

To ensure the security of the stock order information (OI), the OI is encrypted using the PIN input and sent to Hongkong Post’s Certification Authority server for authentication. All transaction details are secured by Hongkong Post Mobile e-Cert, secured by PKI technology. The Mobile e-Cert can be obtained from Hutchison Telecom, the world’s first certified Registration Authority for the issuance of Mobile e-Cert. A high security solution ECC163 cryptography is adopted. Upon verification, the order is allowed to pass through the Order Routing System gateway of Hong Kong Exchanges and Clearing Limited (HKEx) to the designated brokerage firm’s system for subsequent processing. Once the transaction is completed, a confirmation note with a

Databases for Mobile Applications

transaction reference number is sent back to the user’s mBroker interface.

Disaster Recovery in mBroker

Oracle8i database has a component called Data Guard, which offers the most complete and robust disaster recovery solution and high availability through the use of a transactionally consistent standby database. The Data Guard automates the complex tasks of disaster recovery and provides monitoring, alerting, and control mechanisms to maintain a standby operation. Moreover, Data Guard reduces planned downtime by utilizing the standby server for maintenance and routine operations in addition to reporting.

Current Performance of mBroker

mBroker takes less than six seconds on average to complete an order via a handheld, whereas the user takes approximately one minute to complete the same using a phone call. The issue of wireless security is addressed by the introduction of user/server authentication and digital signatures. As a result, an automated, speedy order placement in a highly secured wireless environment is ensured.

FUTURE TRENDS

In view of the issues discussed so far it is obvious that there are many areas which need to be improved. In the future, the goal is to provide the following functionalities for mBroker (Huntsman, 2003):

•Providing intelligent roaming capabilities to enable users to work without interruption, even when network connections are disrupted.

•Exploiting multiple network interfaces in a single device or being able to select the fastest or least costly connection.

•Successfully synchronizing databases by caching contents to local devices through asynchronous connections.

•Allowing portability to a range of devices.

•Conserving power at the operating system level and maximizing performance.

CONCLUSION

In this article we have provided a brief background on the use of databases for mobile applications. This is a grow-

ing area and is facing a number of challenges at this time.

The database design for mobile applications is also dis- D cussed in this article . We have also discussed a success-

ful mobile application called mBroker which is currently in use in Hong Kong. Another similar example is the new mobile workforce effectiveness solution called SMARTselling, developed by Eleven Technology and powered by SQL Anywhere Studio from iAnywhere Solutions. This technology is currently being used by Pepsi Bottling Group and by Proctor and Gamble. This software helps in automated order entry and runs on a small handheld device that communicates wirelessly with backoffice systems. With the help of this application, tedious, error-prone, and costly paper-based processes can be eliminated, and the time spent on checking inventories and shelf displays can be significantly reduced. It is hoped that in the future more mobile applications like mBroker and SMARTselling will be developed, which will affect the various facets of everyday life for people around the globe.

REFERENCES

Barbara, D. (1999). Mobile computing and databases: A survey. IEEE Transactions on Knowledge and Data Engineering, 11(1), 108-117.

Browder, K. (2002). Technical comparison of Oracle database and IBM DB2 UDB: Focus on security. Retrieved December 2, 2003, from http:// www.oracle.com/ip/se/o9idb_db2_techcompar.pdf

Budiarto, K., Harumoto, M., Tsukamoto, M., Nishio, S., & Takine, T. (1998). Replica allocation strategies for mobile databases. IEICE Transactions on Information and Systems, E81-D1, (pp. 112-121).

Budiarto, K., Nishio, S., & Tsukamoto, M. (2002). Data management issues in mobile and peer-to-peer environments. Data and Knowledge Engineering, 41, 183204.

Chan, B. Y. L., Si, A., & Leong, H. V. (1998). Cache management for mobile databases: Design and evaluation. Fourteenth International Conference on Data Engineering, ICDE-98, Orlando, Florida (Vol. 2, No. 7, pp. 54-63).

Dhawan, C. (1997). Mobile computing: A systems integrator’s handbook. New York: McGraw-Hill.

Gok, H. G., & Ulusoy, O. (2000). Transmission of continuous query results in mobile computing systems. Information Sciences, 125, 37-63.

Hung, H.-N., Lin, Y.-B., Peng, N.-F., & Yang, S.-R. (2001). Resolving mobile database overflow with most idle replacement. IEEE Journal on Selected Areas in Communication, 19(10), 1953-1961.

Huntsman, J. B. (2003). Introducing the Intel mobile application architecture guide. Retrieved December 2, 2003, from http://www.intel.com/update/contents/ sw12031.htm

Lai, A., Tam, A., & Lemaitre, S. (2004). Hong Kong stock trading industry embarks on a new era with mBroker capabilities on Palm OS platform. Retrieved December 2, 2003, from http://www.dv net.com/pdf/casestudies/ palm.pdf

Madria, S. K., Mohania, M., Bhowmick, S. S., & Bhargava, B. (2002). Mobile data and transaction management.

Information Sciences, 141, 279-309.

Mazumdar, S., & Chrysanthis, P. K. (1999). Achieving consistency in mobile databases through localization in PRO-MOTION. Second International Workshop on Mobility in Databases and Distributed Systems (MDDS99), Florence, Italy (pp. 82-89).

Munusamy, M., & Hiew, P. L. (2004). Characteristics of mobile devices and an integrated m-commerce infrastructure for m-commerce deployment. Retrieved December 2, 2003, from http://www.wayneyeung.com/files/pa- pers/FP-102.pdf

Prabhu, N., Kumar, V., Ray, I., & Yang, G.-C. (2004, March). Concurrency control in mobile database systems.

18th International Conference on Advanced Information Networking and Application (AINA04), 2, 83-86.

Viellard, E. (2001). Oracle9i Lite Business White Paper. Retrieved December 2, 2003, from http://otn.oracle.com/ products/lite/pdf/o9ilite_bwp.pdf

Yang, X., Bouguettaya, A., Medjahed, B., Long, H., & He, W. (2003). Organizing and accessing Web services on air.

IEEE Transactions on Systems, Man, Cybernetics, Part A: Systems and Humans, 33(6), 742-757.

KEY TERMS

Bluetooth: A wireless technology developed by Ericsson, Intel, Nokia, and Toshiba that specifies how mobile phones, computers, and PDAs interconnect with each other, with computers, and with office or home phones. The technology enables data connections between electronic devices in the 2.4 GHz range. Bluetooth

Databases for Mobile Applications

can replace cable or infrared connections for such devices.

Caching: The technique of copying data from a server machine (the central storage place) to a client machine’s local disk or memory; users then access the copy locally. Caching reduces network load because the data does not have to be fetched across the network more than once (unless the central copy changes).

Database Synchronization: When a database is being synchronized, no new update transactions are allowed, and all open update transactions are finished. After that, all updated blocks are written to disk.

Horizontal Industry Applications: A horizontal industry is one that aims to produce a wide range of goods and services. Horizontal industry applications are utilized across many different industries. While the core part of the application does not require changes, an organization needs customization at the front end or at the back end. Database access and service representative dispatch are typical examples of these applications.

IEEE 802.11: 802.11 refers to a family of specifications developed by the IEEE for wireless LAN technology. 802.11 specifies an over-the-air interface between a wireless client and a base station or between two wireless clients. The IEEE accepted the specification in 1997.

Load Balancing: It is the method of distributing system load evenly across server machines by placing identical copies of frequently accessed information among available server machines.

Middleware: This software manages the communication between a client program and a database. For example, a Web server connected to a database can be considered middleware as the Web server sits between the client program (a Web browser) and a database. The middleware allows the database to be changed without necessarily affecting the client and vice versa.

Mobile Application: A mobile application is any application that can be used on the move. It may or may not be wireless. It must be tailored to the characteristics of the device that it runs on. Limited resources, low network bandwidth, and intermittent connectivity are all important factors that affect the design of these applications.

Mobile Device: A mobile device is anything that can be used on the move, ranging from laptops to mobile phones. As long as the location is not fixed, it is considered mobile. Areas that are not included in the definition of mobile include remote offices, home offices, or home appliances.

Databases for Mobile Applications

Public Key Infrastructure (PKI): A system that enables users of a public network to exchange data securely and privately through the use of a public and private cryptographic key pair, which is obtained and shared through a trusted authority. It provides for a digital certificate that can identify an individual or an organization and director services that can store and, when necessary, revoke the certificates. The comprehensive architecture includes key management, the registration authority, certificate authority, and various administrative tool sets.

Replication: The process of creating read-only copies of any data. Replication is supported by the security, directory, and file services in a distributed computing environment. Replication can improve availability and load balancing.

Secure Sockets Layer (SSL): SSL is a transaction security standard developed by Netscape Communica- D tions to enable commercial transactions to take place

over the Internet. It’s one of a few competing security standards.

Vertical Industry Applications: A vertical industry is one that is focused on a relatively narrow range of goods and services. Vertical industry applications are specific to certain industries. Usually there are some characteristics of the business processes unique to a particular industry that make certain applications very specific for that particular industry. As a result, some vendors develop turnkey software solutions for their own use.