BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Application and data integration definitions
Your most valuable data likely is spread across the company, maintained in different databases that can't "talk" to each other. On top of that, it's changing at a staggering pace, and new data is constantly being added.
Integrating that data is critical to your organization's success, but with so many integration-related acronyms and buzzwords floating around, it can be hard to keep track of what they all mean and what they can do for you. For example, what are EAI, CDC and SOA? And how can odd-sounding terms like "mashups" and "complex event processing" help your company with its data and application integration projects?
To help you brush up on your integration vocabulary, we've compiled a list of eight top terms in the integration market, including extract, transform and load (ETL), enterprise application integration (EAI), real-time integration and more. Each data and application integration definition is linked to useful integration tutorials, training, expert advice and articles to help you optimize integration projects at your organization.
Table of contents
Top eight data and application integration buzzwords and definitions
- Enterprise application integration (EAI)
- Real-time data integration
- Complex event processing (CEP)
- Extract, transform and load (ETL)
- Change data capture (CDC)
- Service-oriented architecture (SOA)
- Data federation
Top application and data integration definitions and buzzwords
Enterprise application integration (EAI) refers to the methodologies and middleware tools used for coordinating applications within an enterprise. EAI often comes into play when organizations update or consolidate their application portfolios. For example, a company may start an EAI project when it's adding or migrating to a new set of applications but still wants to continue using the legacy applications and databases it already has in place. In addition, when companies revamp their business or IT strategies, EAI technologies and techniques can be employed to efficiently tie together new applications with existing ones that fit into the new approach and are still needed.
- Learn about the advantages and disadvantages of EAI and how companies can exploit middleware, such as EAI tools, for their benefit.
In an era where people want the latest and greatest at the snap of a finger, the need for up-to-the-minute data is no exception. As a result, many companies are turning to real-time data integration to make the most current data available to their users. Real-time data integration can be achieved through several different methods, some of which are mentioned later in this list. But the goal is always to transmit accurate and timely data between systems so users can make better-informed business decisions.
- Find out what the different real-time data integration options are and which is the best fit for your company.
- Learn how to avoid enterprise data mashup madness and see how a data warehouse project led to data mashups in Washington, D.C.
Complex event processing (CEP) technology detects and analyzes cause-and-effect relationships in event-based data in real time, so corporate managers and other users can take quick action in business situations . The technology is designed to find patterns in data taken from multiple sources; for example, it often is used by stock traders to identify and react to changes in market conditions and by intelligence agencies to track crime and terrorism suspects. The CEP market is expected to grow by a compound annual rate of 55% from 2008 to 2012, reaching $2.7 billion per year by the end of that span, according to research firm IDC.
- Find out how Informatica hopes to benefit from its recent acquisition of CEP specialist Agent Logic and what the deal means for customers.
Extract, transform and load (ETL) refers to three distinct functions that come together in a single tool and are used to capture and prepare data for analysis, reporting and other uses. First, the extract function pulls data from a source database. Next, the transform function converts the extracted data into a desired form so it can be reused elsewhere. And lastly, the load function writes the converted data to a target database.
- Learn how ETL tools differ from enterprise data replication technology and why open source data integration tools are a good choice for basic ETL jobs.
Change data capture (CDC) is the process of identifying important changes made to the information in data sources, in real time, and then applying the changes throughout an enterprise to ensure that data in different systems remains synchronized. CDC technology minimizes the IT resources required for ETL processes because it only deals with updates and other data changes. It's often used to support data warehouses and operational BI environments, such as call centers where customer service representatives need to be able to access the latest and freshest customer data.
- Read about Oracle's recent acquisition of GoldenGate Software, which specializes in CDC and data replication, and learn how the purchase changes the face of the CDC market.
A service-oriented architecture (SOA) supports communications between two computing services, defining how programs and other IT resources interact with each other and enabling them to perform processing work on behalf of one another. In an SOA, each interaction is self-contained, and databases and applications are loosely coupled instead of being tightly integrated. SOAs support integration efforts by freeing data from applications and making it easier to share and reuse data and common services across a company. Web services based on the Simple Object Access Protocol (SOAP)are a common example of SOA use, although SOAs don't need to involve a Web service.
- Make sure you know the basics of SOA and data integration and how to mitigate the risks and challenges that come with SOA projects.
Data federation technologies pull together data from various source systems to create a single virtual database for BI and other analysis uses. Data federation can be used to complement less agile data warehouses that rely on preset batch-load data transfer methods, or as an alternative to building an enterprise data warehouse. Via data federation, users can amass data from multiple data marts or departmental data warehouses and access information in operational systems for analytic use alongside the warehoused data; the disparate streams of data are integrated, standardized and cleansed in a middleware layer. Data federation is also known as data virtualization, enterprise information integration (EII) and information-as-a-service.
- Learn more about data federation and how data federation technology complements fragmented data warehouses.
Have more integration-related questions that need answering? Browse the Q/As in our integration and SOA expert section with Evan Levy and Jill Dyche. If you don't find the information you're looking for, you can also ask Evan a personalized integration question.
* Some of these definitions were excerpted from definitions on WhatIs.com.
Learn about key products for enterprise application integration