State Of The Data Integration Market 2008-2009

State of the Data Integration Market 2008-2009 An Oracle White Paper November 2008

State of the Data Integration Market 2008-2009

Executive Summary .......................................................................................... 3 Overview of Key Findings............................................................................... 3 A Healthy and Growing Market...................................................................... 4 Regional Adoption of Data Integration .................................................... 5 Data Integration Users Expanding Beyond IT ........................................ 5 Data Integration: A Corporate Imperative .................................................... 6 Business Drivers for Data Integration....................................................... 7 Data as a Foundation for SOA................................................................... 9 Key Obstacles to Data Integration Success ............................................ 10 Key Business Takeaways............................................................................ 11 Data Integration: Functional Capabilities .................................................... 11 Data Movement with Core ETL .............................................................. 12 Data Movement with Next-Generation ELT......................................... 12 Data Synchronization................................................................................. 13 Data Quality................................................................................................. 13 Data Management....................................................................................... 14 Data Governance........................................................................................ 15 Data Services and Data Federation .......................................................... 15 Market Analysis of Functional Categories .............................................. 16 Key Functional Takeaways........................................................................ 17 Three Important Data Integration Trends .................................................. 17 Next Steps for Data Services .................................................................... 17 Real-Time Actionable Business Intelligence........................................... 18 Master Data Management and the Single View of the Business.......... 20 Best Practices: Think Big, Start Small, Act Quickly ................................... 21 Customer Case Studies ................................................................................... 23 Data Consolidation for Retail ................................................................... 23 Data Services for Financial Services ........................................................ 24 Real-Time Business Intelligence for Manufacturing.............................. 26 Oracle’s Data Integration Strategy................................................................ 27 Conclusion........................................................................................................ 28

State of the Data Integration Market 2008-2009 Page 2

State of the Data Integration Market 2008-2009

EXECUTIVE SUMMARY Customer feedback confirms that companies are achieving significant benefits from their data integration, including increased productivity, improved customer loyalty, and lower deployment and maintenance costs.

Data integration is a critical and fundamental element in a variety of technologies, including data warehouses, business intelligence (BI) applications, service-oriented architectures (SOA), master data management (MDM) applications, and datacentric architectures. This white paper reviews the state of the market for data integration for 2008 and 2009. It defines data integration, discusses the trends in the data integration market, and provides lessons learned from real implementations. The information presented is based on the State of the Data Integration Market Survey, a study of more than 300 global data integration users conducted by Oracle in August 2008. It also incorporates industry research and projections by leading analyst firms. OVERVIEW OF KEY FINDINGS

The results from Oracle’s State of the Data Integration Market Survey clearly indicate that the data integration market is growing both in size and corporate importance. It has matured from its extract, transform, and load (ETL) roots to become a top priority for leading companies. Because the business drivers for data integration are so compelling, it remains an investment area even in times of economic stress. Data integration manages the process of combining data from multiple sources into a single, comprehensive view of enterprise data. The ability to transform crossorganizational data from heterogeneous sources into actionable, insightful information has quickly become a competitive advantage for companies that have embraced data integration. The market for data integration includes solutions and services for building, deploying, and managing data warehouses, information systems, and data-centric architectures. Implementing these technologies is critical for companies interested in exploiting the advantages and agility offered by business intelligence and SOA to surpass their competitors and grow market share. To provide context for the ensuing detailed discussion of the data integration market, key survey findings are listed below. The remainder of this document expands on these points to further illustrate the potential value organizations can unlock by creating a unified view of enterprise data.

State of the Data Integration Market 2008-2009 Page 3

•

A healthy, growing, and evolving market. The data integration market is growing at more than 20 percent year over year. We expect this growth to accelerate as data integration becomes needed for data services, BI, and MDM technologies.

•

Underutilization of data within SOAs. Currently, most SOAs are not leveraging data—specifically data services—to their full potential. However, emerging techniques in data federation combined with advanced techniques in data consolidation should create significant growth within both the SOA and data integration markets.

•

Demand for information driving the demand for data integration. Implementing effective BI solutions requires a solid data management foundation. Because companies view BI as one of their most pressing concerns for becoming more dynamic and flexible, the data integration market will flourish.

•

Confusion around MDM. Still in its infancy, the MDM market is fragmented. Experts debate the merits and drawbacks of a domain-specific approach versus a functionality-focused, or domain-neutral, approach. Nevertheless, comprehensive data integration and data management will play a role in redefining the landscape of MDM.

•

Thinking big, starting small, and acting fast. Many best practices for data integration architectures have been derived from implementing SOA. This white paper provides examples of how to keep the data integration process simple while also providing maximum enterprise value.

A HEALTHY AND GROWING MARKET The growth of the data integration segment has exceeded market expectations as companies recognize the fundamental importance of unified enterprise data. The market is expected to exceed US$3 billion by 2012.

The growth of the data integration segment has exceeded market expectations as companies increasingly begin to recognize the fundamental importance of unified enterprise data. Though growth estimates might vary, there is general consensus on the market size of data integration. Two respected enterprise analysts measure the market in excess of US$3 billion by 2012. •

IDC estimates that the worldwide data integration and access software market will grow to US$3.8 billion in 2012, reflecting a compound annual growth rate (CAGR) of 8.7 percent from 2007 to 2012. 1

•

Gartner indicates: “Initial 2007 market share estimates backed by strong fourth quarter and year-end earnings reports across the market landscape indicate solid performance for the worldwide data integration tools market in 2007, with record annual revenue growth of more than 24%.” 2

1 IDC,

“Worldwide Data Integration and Access Software 2008-2012, Forecast doc. #211636,” April 2008. 2 Colleen Graham, “Market Trends, Data Integration Market, Worldwide 2007-2012,” Gartner, May 2, 2008, 4.

State of the Data Integration Market 2008-2009 Page 4

We agree with Gartner's statement about growth in this market, especially considering the explosion of new areas for data integration solutions beyond conventional ETL and into data services and MDM. We further expect high growth rates because data integration is strongly attached to both SOA and BI, both of which are experiencing high growth. In response to volatile global economic conditions, we expect a softening of IT spend for 2009. However, data integration will continue to be a core, fundamental requirement, and we predict that the market will continue to grow as companies rely on data integration initiatives for cost savings and improved efficiency. Regional Adoption of Data Integration

On a regional basis, the data integration market is evolving at predictable levels, with North American and European markets being the first adopters of data integration technology. In addition, our survey indicates a healthy adoption in Asia Pacific.

Which part of the world is your organization located?

43

North America

% Responded  Number of  respondents: 350

28

Europe, Middle‐ East and Africa 22

Asia Pacific 7

Latin America 0

10

20

30

40

50

Figure 1. North America is the leading adopter of data integration technologies.

Data Integration Users Expanding Beyond IT More and more operations staff, data stewards, and business analysts are key stakeholders for data integration solutions. In addition, there has been a spike in the number of IT and enterprise architects that are considering data integration as part of a global enterprise infrastructure.

Evidence of the strength of the data integration market is seen in the growth of the number and type of users. Traditionally, data integration users were DBAs and development staff. Today, more and more operations staff, data stewards, and business analysts are key stakeholders for data integration solutions. In addition, there has been a spike in the number of IT and enterprise architects that are considering data integration as part of a global enterprise infrastructure. This reenforces the belief that data integration has significantly evolved from a simple, low-level, developer-based offering. In the survey, 39 percent of the data integration users were IT or enterprise architects.

State of the Data Integration Market 2008-2009 Page 5

What is your role in your organization? 

39

IT/Enterprise Architect 19

DBA

Number of  respondents: 350

16

IT Management and Operations

% Responded 

15

Business Analysts 9

Data Stewards 4

CXO (CEO,CIO,COO,etc.) 0

10

20

30

40

Figure 2. IT and enterprise architects, operations staff, data stewards, and business analysts are key stakeholders for data integration solutions.

DATA INTEGRATION: A CORPORATE IMPERATIVE

In today’s challenging economic climate, companies are struggling to do more with less; they are retreating to core IT fundamentals and expecting immediate return on their investments. Data integration is no exception. To meet these demanding thresholds, data integration must move beyond ETL to provide enterprise architects with more tangible and compelling benefits. “As organizations increasingly view information-related capabilities as mission-critical parts of the IT landscape, they are moving toward an ‘information infrastructure’ that provides data consistency and interoperability across enterprise applications.” 3

ETL technologies originated more than 10 years ago as an answer to challenges in building, deploying, and managing data-centric architectures. Data warehouses continue to be the key reason for using or evaluating data integration solutions. Oracle’s data integration survey reflects this trend: 55 percent of respondents are either evaluating or using data integration in a data warehouse environment within their organization. Which scenario are you evaluating or using data integration in your  organization today?  55

Data Warehouse 30

Separate standalone ETL platform

Number of  respondents: 350

25

Together with SOA for data services

% Responded as significant

24

Included as part of a BI offering 17

Embedded as part of the databases

Number of respondents: 330

6

Included as part of an MDM platform 0

10

20

30

40

50

60

Figure 3. A data warehouse is the primary reason for using or evaluating data integration, but companies are also considering data integration in conjunction with SOA, BI, and MDM projects.

3

Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 4.

State of the Data Integration Market 2008-2009 Page 6

Business Drivers for Data Integration “As more organizations begin to recognize the role of information management technologies in support of high-profile disciplines and initiatives such as MDM, BI, and SOA, the market for data integration tools will grow.” 4

Unfortunately, simply “making my data warehouse work” isn’t a business driver for companies today. Enterprise architects, together with line-of-business managers, are looking for top-line reasons to justify the IT expenses and organizational changes associated with integrating data. They recognize the role that information management plays in high-profile initiatives such as MDM, BI, and SOA. The drivers for data integration stem from the need to eliminate stumbling blocks associated with turning data into agile information in these high-profile initiatives. In consolidating the research, we identified three categories of business drivers for data integration: improved agility, customer intimacy, and cost cutting. Improved Agility

Accurate, manageable, and transparent data allows organizations to more-quickly identify and respond to internal and external events. Data integration solutions foster this type of agility by uniting heterogeneous datasources across the enterprise. However, these solutions must be well governed to ensure that data is incorporated into business processes and that data integration becomes part of the change management process. Data quality, data profiling, and data governance are essential components to establish and maintain the improved flexibility provided by complex data-centric architectures. “Organizations realize they cannot succeed without a complementary focus on data in their SOA.” 5

Recent data from Oracle’s data integration survey shows that more than 35 percent consider improved agility and improved efficiency as the most significant business drivers for enterprisewide data management initiatives. This is likely due in no small part to the maturation of SOA. Companies now expect their data-centric architectures to be as flexible and service enabled as their event, message, and process contemporaries. Customer Intimacy

Improving the customer experience is one of the leading drivers for data-centric architecture initiatives. Understanding how customers behave by improving data consolidation across organizational units, channels, and partners is not an easy task; it requires that data provide a single version of the truth. Data integration is one of the key enablers for improved customer intimacy. When customer data is clean, consistent, and up to date, it dramatically improves the ability for a company to deliver a high-quality, seamless customer experience. Cost Cutting

The cost of data integration is directly related to the volume of data. More storage disk space, more CPU cycles, and more time are needed to manage rising amounts

Colleen Graham, “Market Trends, Data Integration Market, Worldwide 2007-2012,” Gartner, May 2, 2008, 3. 5 Ibid., 5. 4

State of the Data Integration Market 2008-2009 Page 7

of data. Because data volume is continually increasing, managing cost becomes increasingly important. “With typical investment in data integration tools falling in the range of US$200,000 to US$500,000 for software licensing and US$50,000 to US$100,000 for

While cost cutting continues to be a key driver for data integration, the survey also found that companies are willing to make the required investment before realizing any return on investment (ROI). However, there are several issues that increase investment costs and may inhibit ROI, including

annual maintenance, organizations can achieve software cost savings of

•

Overly complex data architectures and data integration processes; survey results highlight complexity as the primary barrier to data integration success

•

Risk associated with inaccurate and inconsistent data

•

Code-based, SQL, or manual approaches to building data flows

•

Redundancies in data flows

•

Consolidation of data structures and data marts

•

Data tools and database management system (DBMS) consolidation

US$250,000 or more through consolidating tools or replacing tools with lower-cost options.” 6

Without question, information agility, customer intimacy, and cost and risk reduction are clear drivers for data integration. The challenge remains how to educate and align the key stakeholders—business users, architects, IT managers, and data stewards—to consistently and effectively apply these business drivers across the enterprise.

What is the biggest benefit you are expecting from  launching enterprisewide data management initiatives? 35

Improved Agility, improved efficiency 30

Improves customer experience

% Responded  Number of  respondents: 350

22

Cost savings from consolidation 13

Regulatory Compliance mandates 0

10

20

30

40

Figure 4. Information agility, customer intimacy, and cost reduction are clear drivers for enterprisewide data integration projects.

Ted Friedman et al., “Cost Cutting in Data Management and Data Integration,” Gartner, February 15, 2008, 3.

6

State of the Data Integration Market 2008-2009 Page 8

Data as a Foundation for SOA

Companies initiating projects that require data must make choices. Should they continue to connect data in the same customized, rigid, point-to-point way that they use for applications? How can they ensure data is consistent, accurate, and current? How can the data be managed, tracked, and profiled? Experienced IT professionals know that all of these factors are important and need to be considered together. They recognize that custom code or single-use data integration projects are neither scalable nor reproducible, and that they offer negligible ROI. The better strategy is to apply reusable principles to data integration—turning data into a service that is available as logical modules, each with a standards-based interface. This allows users to access and use the service more easily, improves data visibility, and promotes greater reuse. Data reuse and flexibility is one of the key architectural requirements for many large enterprisewide data-centric architectures.

Data reuse and flexibility is often one of the key architectural requirements for many large enterprisewide data-centric architectures. Recent data from Oracle’s data integration survey shows that more than 50 percent of all companies leverage data in the form of reusable data services. This indicates that the importance of data increases as companies focus on enterprisewide SOA implementations. That data integration is a critical enabler of SOA success suggests that the data integration market will grow in conjunction with the increase in SOA investments. Similarly, it implies that organizations that have begun SOA initiatives without considering data integration have a lower chance of success.

Are you leveraging data as part of your SOA today?

% Responded  59

Yes

Number of  respondents: 350

41

No

0

20

40

60

Figure 5. More than 50 percent of companies leverage data as part of a SOA implementation.

State of the Data Integration Market 2008-2009 Page 9

Key Obstacles to Data Integration Success The biggest barrier to data integration success is fragmentation that results from applications and solutions that do not work together. As a result, most of the effort behind data integration initiatives involves building maps between known

The biggest barrier to data integration success is fragmentation. In many cases, today’s applications do not work well together out of the box. In addition, these applications do not work with common SOA or BI platforms. In other cases, companies deploy multiple, varied solutions across different divisions, and these solutions might not integrate. As a result, most of the effort behind data integration initiatives involves building maps between known systems that are not integrated.

systems that are not integrated.

Figure 6. Fragmented solutions are common in large enterprises.

In the survey, more than 58 percent of the respondents cited multiple fragmented solutions as the leading obstacle to successful implementations. These results map directly to the findings of several industry analysts. Custom coding was cited as the second most significant obstacle, with performance a close third. Unfortunately, many conventional ETL approaches require custom coding as the basis to tune or configure their deployments. ETL has for many years been the business enabler that consolidates disparate data while moving it from multiple datasources; thus, this customization is unlikely to disappear soon. In addition, with the growth of data continually on the rise, performance and scalability will likely be key requirements for all new initiatives. What are the impediments to successful data warehouse  implementations? Multiple and fragmented solutions across the organization

59 38

High cost of maintaining custom code

% Responded as significant

37

ETL Performance issues

Number of  respondents: 350

30

Lack of adoption or funding across projects

27

IT and Business not aligned ETL tools are hard to use 

23 0

20

40

60

Figure 7. Almost 60 percent of respondents to the Oracle survey cited fragmented solutions as an impediment to data warehouse implementations.

State of the Data Integration Market 2008-2009 Page 10

Key Business Takeaways

Data integration is becoming more important to corporations because information can provide differentiation and a competitive edge. Business drivers for data integration are focused on leveraging information more efficiently; these tend to translate to lower costs, customer intimacy, and improved agility. Meanwhile, IT demands accurate, up-to-date data that is highly accessible and flexible. Achieving these objectives can be challenging. The fragmented nature of enterprise applications becomes a significant obstacle and is one of the most significant issues facing organizations that want to integrate their data. By adopting a solution approach to data integration, companies can avoid the pitfalls of fragmentation. Such an approach unifies many of the functional capabilities of data integration into a cohesive platform. DATA INTEGRATION: FUNCTIONAL CAPABILITIES By definition, a comprehensive data integration solution must provide data movement, data synchronization, data quality, data management, and data services capabilities.

Although data integration functionality can differ somewhat depending on classifications of the business and IT problems addressed, there is general agreement that a data integration solution includes five core functional capabilities: •

Data movement. Core ETL capabilities, bulk data transfer.

•

Data synchronization. Change data capture, data replication.

•

Data quality. Data cleansing, data quality business rules.

•

Data management. Metadata management, MDM, data modeling.

•

Data services. Single and federated data access, bulk data services.

Each customer can prioritize the functional capabilities above in a different manner, depending on the intended scale of their data integration deployment, organizational skills and preferences, and types of technologies already in use.

Figure 8. A data integration solution is composed of data services, data management, data quality, data synchronization, and data movement capabilities.

State of the Data Integration Market 2008-2009 Page 11

Data Movement with Core ETL

ETL technology supports the extraction, transformation, and integration of data from multiple datasources, including databases and data warehouses. ETL allows businesses to consolidate their disparate data from any source while moving it from place to place. ETL can transform not only data from different departments but also data from different sources altogether. For example, order details from an enterprise resource planning system and service history from a customer relationship management application can be consolidated into a central data hub for a single view of the customer. Although ETL technology is still heavily used for data warehousing and BI initiatives, data and knowledge management professionals are increasingly demanding additional data integration capabilities from their ETL vendors to support complex data integration challenges. Data Movement with Next-Generation ELT Solutions that optimize where data transformation is performed are known as E-L-T—or extract, load, and transform— tools because they can optimize where transformations are deployed.

Solutions that optimize where data transformation is performed can distinguish themselves from conventional ETL, or extract, transform, and load, approaches. These solutions are known as E-L-T—or extract, load, and transform—tools because they can optimize where transformations are deployed either on the target destination or even on the source. This also allows for greater flexibility, improved scalability, and greater performance. ELT approaches can also reduce costly IT infrastructure costs. The following should also be considered when evaluating an ELT solution: •

Performance optimizations for set-based transformations.

•

Heterogeneous relational databases.

•

Optimizations for database appliances.

•

No hardware requirements; run-time agents should be deployed on the databases themselves.

•

Data that always goes from source to target through optimized database pathways; data should never move through the intermediary.

•

Extensible support for standard Java and SOA environments

•

Design tools that support out-of-the-box optimizations. Users should not have to write special scripts or custom code to enable optimized performance.

Not all ELT approaches are equal. When selecting an ELT solution, it is important to discern between brittle proprietary technologies that can easily break and open ELT platforms that can dramatically improve performance while simultaneously lowering cost of ownership.

State of the Data Integration Market 2008-2009 Page 12

Data Synchronization

There are many ways to extract data from a DBMS, including queries, replication, table dumps, storage snapshots, and calls to the API of an application that sits over the database. Change data capture (CDC) is an alternate data extraction method that has recently become of interest, primarily because it enables data integration to operate closer to real time. CDC can be applied to most database brands, including relational, legacy, mainframe, and file-based DBMSs. A few vendors have built CDC into their products, but many organizations use the data modeling and log capabilities of a DBMS to build their own solutions. CDC has been around for many years, but its ability to solve some of the most difficult data integration challenges is driving interest among IT professionals today. A simple example of CDC in action follows. Two separate datasources for a web storefront (one for customer data, one for order data) are consolidated into a single data warehouse. To simply update the order details in real-time, only the delta (or set of orders and new customer info) needs to be propagated across to the data warehouse. This does not require moving all the data for both systems. Without CDC, business managers would not be able to see daily trends. In addition, business managers would be forced to wait for the next batch of data to load into the data warehouse before they could look at the results. By then it might be too late to make important informed decisions. Data replication is another key component within synchronization technology that is required in any effective core data integration offering. It is a distinct requirement from CDC in that it is often needed in deployment considerations for mirroring or maintaining identical data across data centers. CDC is required for synchronizing data across heterogeneous datasources, whereas data replication technology is often embedded within database tools or data warehousing tools. Data Quality All enterprise software projects are at risk from bad data; even worse, inaccurate and inconsistent data exists everywhere. However, the demand for trusted data continues to increase, driven by investments in packaged applications and BI software.

All enterprise software projects are at risk from bad data; even worse, inaccurate and inconsistent data exists everywhere. However, the demand for trusted data continues to increase, driven by investments in packaged applications and BI software. Strategic IT initiatives such as MDM also add additional pressure. Further complicating the matter, regulatory compliance initiatives—such as the SarbanesOxley Act, the U.S. Patriot Act, and Basel II—require tracing the source of the data used in financial reports, as well as examination, tracking (through snapshots), and certification of the state and quality of the business data. The act of data profiling is often omitted when trying to achieve data quality. Data profiling is a data investigation and quality-monitoring mechanism that allows business users to assess data quality through metrics, discover or infer rules based on this data, and monitor the evolution of data quality over time. Data profiling

State of the Data Integration Market 2008-2009 Page 13

works with data quality to better understand and manage the holistic issues associated with data quality. According to the Oracle survey, more than 83 percent of customers do not have complete trust in their data. How much do you trust your data today?

35

Few non‐critical issues

% Responded  30

Average issues with data cleansing and quality of data Regular data gaps which require manual intervention

Number of  respondents: 350

18 17

100% Confidence 0

10

20

Number of respondents: 330 30

40

Figure 9. More than 80 percent of respondents indicated that inaccurate and inconsistent data exists in their organizations.

Data Management “Too many organizations have left their data unmanaged in several critical ways: lack of standards, loose or nonexistent

Data management is often the forgotten requirement in a data integration solution, but it as important as the other components. At the core of any data management solution is the need for metadata management, MDM, and data modeling.

policies for security and access control, minimal focus on data quality, and significant fragmentation and redundancy.” 7

Metadata management improves data visibility so managers can understand how data is used and how it relates to other data within a global data-centric system. Metadata management and data relationship management are cornerstones for MDM-based solutions that reveal data relationships within a single source of truth. Data lineage is a key example of metadata management often used by BI utilities to allow business users to independently track datasources. If the data lineage falls short of the actual source and has not integrated properly to the data integration solution, it will be unable to allow business users to identify gaps in the data. Data modeling is a key element to the design aspect of creating and describing information architectures. Data integration solutions are not always packaged with out-of-the-box data modeling capabilities. However, a data integration solution needs to integrate with data modeling capabilities for successful data management. This is especially true when leveraging aspects of data management that change the data model as part of the management activity.

7

Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 5.

State of the Data Integration Market 2008-2009 Page 14

Data Governance

Investing in technology alone does not deliver trusted data. Information managers need to define what data means to their organizations through data governance. Data governance is analogous—though not identical—to SOA governance, a discipline that has saved many SOA implementations from certain failure. “Governance, regardless of what specialty you are talking about, is a series of activities associated with influencing the actions and behavior of an environment. SOA governance, data governance, process governance, and application governance are all related to, but not dependent on, one another. It's not that one encompasses the other, but rather the activities associated with each should work in conjunction with the rest of the governance discipline.” 8 Data governance helps define not only data quality rules but also the processes for how the rules are maintained, approved, and iterated. As companies scale and grow, these established processes are critical to managing the lifecycle of enterprise datacentric architectures. Data governance must include multiple data quality and data management capabilities, as well as allow for the human element in implementing a governed data-centric environment. For example, a company might define certain data as off-limits to a set of roles that is integrated across multiple data hubs. This type of governance can be implemented by combining identity management and data access services or through an entitlement policy that is executed at runtime. In other cases, data quality might require a complex set of business logic to be specified as a business rule, or business processes might automate a workflow of data exception management. In each example, governance processes are key to successful enterprise implementations. Data Services and Data Federation Data services are the foundation of many SOA deployments and are needed to bridge the gaps between processes and the core application infrastructure.

Data services have a transformational influence on enterprise data-centric architectures. Data services are the foundation of many SOA deployments and are needed to bridge the gaps between processes and the core application infrastructure. While there are many categories for data services, Data access services are the most commonly used. Our analysis indicates that there are three important scenarios where data can be exposed as reusable access services. •

Single data access

•

Data hub data access

•

Data federation services for multi-source data access

Michael Stamback, “SOA Governance and Data Governance: Related or Distant Cousins?,” blogs.oracle.com/governance, September 2008.

8

State of the Data Integration Market 2008-2009 Page 15

Many off-the-shelf products in SOA, BPM, databases, and even development tools include basic functionality for accessing datasources at a single level. Consolidating data into a hub using ETL and CDC, and then building real-time access points as services is another viable option. However, virtually aggregating data access services from multiple heterogeneous sources is more challenging. This scenario is known as data federation. Data federation leaves data at the source and consolidates information virtually—in a manner very similar to how an enterprise service bus virtualizes messages. Data federation allows companies to aggregate data across multiple sources into a real-time view that can be reused as a service. When there are restrictions for accessing data at the origin rather than at the data hub, this technique—combined with data consolidation approaches—is especially useful. 9 The Oracle survey revealed that more than 31 percent of data integration users realize that real-time data access is the most pressing concern for the technical evaluation of data integration solutions.

What is the single most important component you are looking for  in a data integration solution? % Responded 

31

Real‐time data access

Number of  respondents: 350

19

Data Migration 17

Data Services

16

Data Quaity and Profiling 11

Data Movement 6

Federated data access 0

10

20

30

40

Figure 10. One-third of data integration users believe that real-time data access is the most important component for a data integration solution.

Market Analysis of Functional Categories “Efforts to reduce the total cost of ownership will drive businesses to consolidate on one tool or vendor that increases operational efficiency and enables better use of data management resources. Already, many organizations are are standardizing on a single data integration tool or set of tools. As a result, larger vendors, including DBMS vendors,

Initial estimates indicate that data services have only captured 5 to 10 percent of the data integration market while its predecessor, ETL, still holds the largest share of the market. However, the emerging data services trend has already had a transformational influence on enterprise data-centric architectures. In addition, data quality, data management, and data federation are becoming increasingly popular, but they still are not mainstream technologies in the overall market. Going forward, more and more solutions will likely take advantage of these emerging elements to capitalize on the benefits of real-time information approaches.

will benefit most from the growth of the data integration market.” 10

9

Dain Hansen, “Demystifying Data Federation,” SOA Magazine, August 2008. Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 26.

10

State of the Data Integration Market 2008-2009 Page 16

Key Functional Takeaways

Businesses need to consolidate their disparate data and move it from place to place so that it can be used by the appropriate applications. As a result, the need to transform and move data has been the leading reason for adopting data integration tools. Together with data synchronization, these data hub approaches enable data integration to operate closer to real time. However, without data quality, data management, and data services, these data hubs become point solutions that are difficult to maintain and reuse. We expect companies will begin to combine all these functional capabilities—data movement, data synchronization, data quality, data management, and data services—within a single data integration solution. Such comprehensive solutions can enhance the benefits of SOA, BI, and MDM initiatives and achieve maximum business value. THREE IMPORTANT DATA INTEGRATION TRENDS

Technical innovation is at the heart of the data integration market growth. Recent advancements make it easier than ever before to create a nimble, distributed application and system architecture while still harvesting information from an array of datasources in a meaningful way. Our research indicates three specific business trends that are being exploited and enhanced through data integration functionality: •

Data services

•

Business intelligence

•

Master data management

Next Steps for Data Services

Many enterprises see data integration as a key element to their SOA. When SOA is implemented to fix integration problems, data integration is almost always affected. As a result, a clear trend in data services emerges: data services will likely follow the mainstream adoption and successful momentum of SOAs. There is interesting discussion on the direction for data services. Some industry analysts believe data services will evolve into multiple types: enterprise search, reporting, and a single view of the truth. Others admit that they are skeptical about the long-term viability of data services. The skepticism is typically a result of the performance challenges inherent in delivering large-scale federated views. Another concern relates to maintenance. According to Gartner, “Disillusionment will occur as organizations realize the wide variation in services, ranging from possibly thousands of very fine-grained services performing rudimentary tasks through coarse-grained and composite services.” 11

11

Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 14.

State of the Data Integration Market 2008-2009 Page 17

“To achieve the business benefits expected from SOA initiatives, organizations must include a strong dataoriented perspective in their work. Focusing only on the modularization, reuse, and composition of application logic is not enough—SOAs also demand that firms address data issues (such as poor data quality and lack of consistency in semantics across applications and data stores).” 12

At Oracle, we believe that the term data services will expand beyond simple data federation—or, multisource data access—and begin to incorporate data management. This expansion of the data services definition will occur as vendors better integrate support for data services into their data integration solutions. Indeed, individual vendors are already beginning to merge these data integration technologies into a single solution, offering multiple tools that can be used at will. As federation becomes useful for BI and MDM applications that demand flexible and agile data integration, this approach will become more common. And when enterprises begin to build out hybrid data-centric environments that include both data hubs and data aggregates, they will turn to new types of data governance offerings for data services. Overall, we believe that data services will ultimately succeed. Companies evaluating data integration strategies should consider data services for multisource aggregation in the short term. In the long term, data services will likely be part of overall data integration and data management strategies. Real-Time Actionable Business Intelligence

“The typical BI framework consists of a data warehouse with reports, but the answers to some questions will always lie outside that data warehouse. Users need access to broader sources of data through data integration and federation; therefore, BI leaders should create a much broader framework.” 13

Increasingly, companies rely on BI systems for mission-critical decisions and planning. However, many of today’s conventional BI solutions are simply batch transfers of operational data into data marts that serve up the data in dashboard applications and batch-generated reports. In addition, BI systems are only as good as the data they present. Often a BI implementation will suffer from information inaccuracy, obsolete data, or out-of-sync warehouses. And most data integration tools in the market today suffer from lack of interoperability with BI systems. Because decision-makers are demanding more real-time visibility, actionable recourse, and flexibility in data analysis, this approach to BI is no longer sufficient. Oracle’s survey showed that more than 60 percent of companies now realize that data integration enhances the value of BI applications by increasing the quality and consistency of data. Further, the survey revealed that nearly 53 percent of enterprises believe data integration is a key enabler for BI applications to connect to more systems.

Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 5. Kurt Schlegel, “Q&A: Seven Questions Every Business Intelligence Leader Must Answer,” Gartner, April 9, 2008, 4. 12 13

State of the Data Integration Market 2008-2009 Page 18

How is data integration critical for your  business intelligence initiatives? 60

Helps BI applications have better quality and consistent data 53

Helps BI applications connect to more systems

Number of  respondents: 350

38

Allows BI applications to behave in real‐ time 33

Enables better business responsiveness. 0

20

40

% Responded as significant

60

Number of respondents: 330 Number of respondents: 330

Figure 11. 60 percent of respondents recognize that data integration improves the value of BI applications.

To understand how data integration fits within the BI landscape, look at detailed examples where applications consume real-time data and turn it into in-depth analytics and information for improved decision-making. In many such scenarios, change data capture (CDC) plays a key role in keeping data consistently updated without impacting the target or source performance. In addition, these systems draw from a wide range of internal sales, customer, and financial data applications as well as third-party systems. This requires a broad range of data integration connectivity options to support moving data across such a wide variety of enterprise applications. Data quality is important for BI applications, but for different reasons than data warehousing initiatives. DBAs think of quality differently than a business user using a BI tool. DBAs care about the semantics of data, such as broken validations at the field level, whereas the business user considers holistic patterns or matches in the context of customer or performance data. A key offering in the data integration market is actionable BI, which allows users to view and change data based on business processes.

Another key trend in the data integration market is actionable BI. Visibility into data is no longer sufficient; users must be empowered to directly act on this data based on the available information. For example, a BI solution might report that a partner is no longer meeting a service-level agreement, but how does the company act on that data? The partner must be demoted from a platinum-level profile to a lower-profile category. This in turn means changing the data in the data warehouse, or “closing the loop” among data integration, business process integration, and business intelligence. It is an important trend as more solutions take advantage of interoperability points in SOA, BI, and data warehousing. To achieve real-time, actionable BI, the business user must be able to easily drill into the data behind a dashboard to see details about the data lineage—that is, where the data came from and what transformations were applied. When a BI process cuts across data silos, the project team needs metadata to understand the context of information, including terminology, calculations, and methodologies—all prerequisites for a single version of the truth. Achieving actionable BI relies on

State of the Data Integration Market 2008-2009 Page 19

accurate data throughout the data lifecycle—from data origin to data retirement. Data integration is a critical enabler of this data integrity. The Oracle survey revealed that nearly 47 percent of respondents believed business and IT alignment issues were the biggest stumbling blocks to successful BI initiatives. This problem can be resolved when IT meets business needs with a comprehensive data integration solution that enables business processes to take advantage of corporate data.

What were the stumbling blocks to successful  business intelligence initiatives? % Responded as significant

47

Business and IT alignment issues

Number of  respondents: 350

40

Limited penetration to data 36

Traceability and visibility of data lineage

33

Need for more real‐time data access

31

Challenges with multiple solutions 0

20

40

60

Figure 12. Almost half of respondents to the Oracle survey believed that business and IT alignment issues hinder the success of a BI initiative.

The strongest BI offerings embed versatile data integration solutions that increase the value of the information delivered to the business user. Optimizing data integration within a BI solution delivers consolidation across complex applications, clean and consistent data, real-time data access, and actionable BI. Master Data Management and the Single View of the Business “A holistic MDM strategy is required to create a single view of the business, to provide consistent processes and internal reporting within the organization, and to optimize collaborative processes with partners in the supply chain. Data integration tools are key to meeting these needs because they support the

One of the most significant areas of debate in the data integration market involves master data management (MDM). There is disagreement about the role that MDM plays in managing data-centric applications and the future role that it has in redefining data integration platforms. What does is it mean to master your data? Why is it essential to consider as part of a data integration platform? MDM consists of two approaches: one that addresses domain-specific needs and one that is domain agnostic.

requirement to consolidate, integrate, and

“Historically, MDM products were designed to focus on particular MDM scenarios. For example, customer data integration (CDI) hubs focused on the customer data domain, while product information management (PIM) products focused on the product data domain and tended to be more workflow-oriented and collaborative in nature. A separate thread of MDM

synchronize master data across applications and data structures within the enterprise.” 14

Colleen Graham, “Market Trends, Data Integration Market, Worldwide 2007-2012,” Gartner, May 2, 2008, 5.

14

State of the Data Integration Market 2008-2009 Page 20

development focused on “downstream,” analytical MDM requirements and tended to be more data-domain agnostic.” 15 Domain-specific MDM addresses a domain such as a single view of product, customer, supplier, site, or financial data. There are multiple modes for these MDM domain models depending on the industry or the requirement. In each of these domains, data integration is implicitly required, but not necessarily extensible for other purposes beyond the realm of the domain. With domain-agnostic MDM, there are functional capabilities that relate directly to components found in data integration platforms. These include data movement, data synchronization, data quality, data federation, and especially data management, which take into consideration metadata management. This approach masters data for any domain—often seen as a single view of the truth. In this definition, MDM includes platform capabilities for creating “a single view of the business, to provide consistent processes and internal reporting within the organization, and to optimize collaborative processes with partners in the supply chain.” 16 These MDM platform approaches require comprehensive data integration capabilities to ensure that all parts of the enterprise cooperate and work toward common goals. “Software alone won't support MDM. Although many vendors address various parts of MDM, no single vendor offers everything needed to implement MDM fully at all levels. As a result, MDM initiatives require constant attention. IT leaders must set up a vigorous governance system to ensure that all parts of the enterprise work toward the same goal.” 17

Regardless of the approach, MDM as a platform is still in its infancy. A single view of the truth for all enterprise data is almost viewed as a luxury. Today’s companies continue to struggle in their MDM initiatives because most vendors have yet to deliver unified, comprehensive MDM solutions that combine both domain-specific and domain-agnostic aspects. In fact, platforms require significant customization and professional services to ensure a successful MDM implementation. As a result, enterprise architects and data stewards should exercise caution before undertaking an MDM strategy without first implementing core data integration solutions that integrate their data-centric applications. Despite the chaos and uncertainty in the diverging MDM definitions, data integration can be seen as the cornerstone for successful data-centric architectures and provide authoritative master data for a single view of business. BEST PRACTICES: THINK BIG, START SMALL, ACT QUICKLY

For CIOs, enterprise architects, data stewards, and IT managers working diligently to deliver a better service experience with fewer resources, we offer five simple recommendations based on our research and the findings from the Oracle survey. Although every organization is unique, these recommendations focus on translating business requirements into pragmatic approaches and strategies that are generally applicable to most situations.

Eric Thoo et al, Hype Cycle for Data Management, 2008, Gartner, July 9, 2008, 26. Ibid. 17 Kurt Schlegal, “Q&A: Seven Questions Every Business Intelligence Leader Must Answer,” Gartner, April 9, 2008, 7. 15 16

State of the Data Integration Market 2008-2009 Page 21

•

Avoid fragmented solutions. In our survey, fragmented solutions are cited as the No. 1 reason for failed approaches to data warehousing. Currently, many vendors offer solutions that are not integrated into broader SOA and BI offerings. We recommend looking to solutions that are hot-pluggable to existing IT infrastructure and that support industry standards, such as SOA and Java.

•

Solve business and IT alignment issues first. Alignment issues between business and IT groups were one of the primary inhibitors of successful BI deployments. BI requires alignment from both sides, so this is not very surprising. Sometimes technology alone doesn’t solve alignment, and groups need to properly coordinate their business objectives with IT enablement; for example, communicating what data needs to be reflected in a BI dashboard can uncover design gaps. Perhaps the data for a particular content area needs to be consolidated before it is available for the dashboard. Through proactive communication and cooperation, these issues can be uncovered early in the development process, saving resources while expediting implementation.

•

Consider data governance early in the process. One of the critical mistakes that companies make is building a project around a one-time effort to do data integration. They neglect to develop any governance processes until it is too late. As such, continuous data management and data quality are omitted. How can data continuously be made available, clean, and in sync? Enterprise architects and IT managers need to consider both architecture and business strategies that encourage continuity and forward-looking design. Governance processes—and not just the technology alone—are needed to manage complexity from the outset and stop problems before they arise.

•

Implement data services for improved agility. Despite SOA’s popularity and the agility that it provides, SOA doesn’t require that every data field be exposed as a service. Federated data services offer advantages as well, but are not necessarily the right tool for every data integration initiative. Performance, manageability, and security could override agility and flexibility. These are important considerations and need to be weighed accordingly.

•

Start small, show incremental value, and repeat. One of the recent lessons learned from SOA implementations is to start projects on a smaller scale— despite the urge to cross enterprise boundaries for immediate agility benefits. The same lesson applies to larger data warehouse, MDM, and BI projects that expand in scope across the company. The most successful data integration projects are ones that solve a manageable problem that exists across the organization, while still providing incremental value to the business. For example, using data services enables the incremental reuse of information by the processes and applications that need them for a particular project, while leaving existing infrastructure in place.

Enterprise architects and IT managers need to consider both architecture and business strategies that encourage continuity and forward-looking design. Governance processes—and not just the technology alone—are needed to manage complexity from the outset and stop problems before they arise.

State of the Data Integration Market 2008-2009 Page 22

CUSTOMER CASE STUDIES

To put these best practices into context and provide evidence to support our recommendations, we provide three customer examples across distinct data integration opportunities: data consolidation, real-time business intelligence, and data services. Data Consolidation for Retail

A retail enterprise was suffering from costly errors between its distribution centers and its retail chains. The company realized that data dropouts and reconciliation issues translated into substantial business risk. For example, valid SKUs (or stockkeeping units) on one system would be invalid on another due to out-of-sync data. The objectives for this data consolidation implementation in the retail industry were to gain a better understanding of how data was flowing into multiple fragmented systems, be able to manage data quality exceptions as part of a workflow, and be able to manage data movement as part of a process.

The objectives for the implementation were to gain a better understanding of how data was flowing into multiple fragmented systems, be able to manage data quality exceptions as part of a workflow, and be able to manage data movement as part of a process. The methodology invoked data movement, managing data exceptions through existing process management and SOAs, and leveraging data services as part of a Business Process Execution Language process flow. The key issues facing the implementation team included •

Data errors. Data was unclean. The customer was constantly plagued by invalid SKUs. Because the systems were not integrated, there were constant mismatches that caused process breakages and slowed replenishment.

•

Fragmented data. Many systems were integrated through complex code and custom scripts, so it was difficult to reconcile data across multiple systems. This was especially the case where data was not updated in the inventory system, but was reflected in a separate ordering system.

•

Broken processes. The majority of processes were point to point, making them brittle and difficult to change. These hardwired data access points led to challenges when the system needed to be updated.

•

Lack of data reuse. The enterprise included hundreds of retail chains, multiple warehouse and distribution sites, and several data centers. Yet all of them were using different processes and sharing data in different ways. This lack of consolidation and lack of a shared approach created a huge business risk and ultimately slowed down many business processes.

•

Limited visibility. The business lacked centralized organizational views, activity monitoring, and BI dashboards; business analysts had no way to fix problems or iteratively improve them.

State of the Data Integration Market 2008-2009 Page 23

Figure 13. The data integration solution for this retail customer connected applications with datasources.

When the comprehensive data integration solution was implemented, it connected the retail company’s heterogeneous applications and datasources together so that it was able to

By implementing a robust data integration solution, the retail enterprise could eliminate the cost of errors and focus on optimizing business processes and getting better informational analytics from the clean data.

•

Ensure valid product data and eliminate mismatches or data gaps

•

Optimize order-to-receive processes and ensure consistent data sharing

•

Enable clear visibility of all information, orders, and replenishment activities

•

Shorten reaction time to orders, replenishment cycles, error conditions

•

Eliminate costly mishandlings of orders and replenishment

By implementing a robust data integration solution, the retail enterprise could eliminate the cost of errors and focus on optimizing business processes and getting better informational analytics from the clean data. Only after first integrating data could the company embark on BI and activity monitoring and see some of the early results from SOA and BPM. This is one of the key reasons that data integration is a key foundation enabler for BI and SOA initiatives and continues to be a key corporate imperative. Data Services for Financial Services

One global financial organization struggled with the continuous flux of merger and acquisition activity in which data integration was a perpetual challenge. In this example, disparate subsets of information prevented the organization from achieving a single, organized customer view. This impacted its customer intimacy

State of the Data Integration Market 2008-2009 Page 24

and slowed the development of improved financial service products. Because customer insights were fragmented, the organization’s responsiveness to customers diminished as well. The organization was challenged to correct these gaps by establishing a data services layer to reconcile data, alleviate the demand on DBAs and developers, and reduce time to value of new financial service offerings. This organization had started down the path of simple data access using homegrown mechanisms but quickly realized that the solution wouldn’t scale, was hard to manage, and didn’t perform. In addition, the organization hoped to leverage its data tier and existing data hubs that had already consolidated most of the company data.

Figure 14. By federating data within a Web services layer, the financial services company was able to improve its view of the customer.

This financial services organization was challenged to improve its customer relationships by establishing a data services layer to reconcile data, alleviate the demand on DBAs and developers, and reduce time to value of new financial service offerings. In addition, the organization hoped to leverage its data tier

The objective of the data services solution was to develop a highly flexible, scalable, and single view of the customer based on data across fragmented datasources. The methodology required federating data within a Web services abstraction layer for better views of customer information. It leveraged the existing data mart as the source. The key issues facing the implementation team were as follows: •

Physical changes to the databases needed to be abstracted out so that they didn’t impact the consumers of the Web service.

•

The data services needed to be implemented as an abstraction, not as a replacement of existing data marts and data hubs.

•

Data services were required to scale without performance degradation.

•

The data services solution needed to preserve or improve on current servicelevel agreements (SLAs).

and existing data hubs that had already consolidated most of the company data.

State of the Data Integration Market 2008-2009 Page 25

As a result of implementing data services using the federation technique, the financial services company reduced overall data complexity while increasing data consistency and reusable data services.

•

The original data services had originally been ineffective because each line of business treated customers separately, with different interface requirements. Multiple customers and partners needed to be supported in a single interface.

•

Other business units had refused to adopt the original Web services interfaces because the data implementations were rigid and inflexible. These new implementations needed to be agile, quick to implement, and conform to the business requirements of different groups.

As a result of implementing data services using the federation technique, the company reduced overall data complexity while increasing data consistency and reusable data services. These services initially were extensions to the data tier implementation and later were adopted as new models within the data warehouse. This type of hybrid model reduced development time from 12 weeks to 80 hours; managing and maintaining the implementations now requires half the number of development resources. No changes to the database were required, freeing the DBA teams to address other tasks. User access to data no longer required communication with back-end systems, preserving performance within SLA restrictions. The ROI extends beyond this individual project, freeing this growing organization from constraints that might affect customer relationships and future business. Real-Time Business Intelligence for Manufacturing

A leading manufacturing company in the semiconductor industry had grown significantly, both in sales and size. The company was using tools such as Microsoft Excel and Microsoft Access for budgeting and reporting and was grappling with the inefficiencies of using unscalable tools in a rapidly growing company. So much time was spent aggregating and extracting data that a thorough analysis of the business could not be conducted.

Figure 15. For this manufacturing company, implementing a BI application also meant providing data integration and data quality so that business users made decisions based on the right information.

State of the Data Integration Market 2008-2009 Page 26

A growing manufacturing company implemented a BI solution with the goal of achieving near-real-time analysis of data and reducing the amount of time spent aggregating and extracting data. The methodology used required the company to implement data integration and data quality in conjunction with the BI solution.

The objective of the company’s BI project was to achieve near-real-time analysis of data and drastically reduce the amount of time spent aggregating and extracting data. Data quality is imperative for BI applications. Without it, business users build insight and reports on the wrong information. Hence, the methodology used required the company to implement data integration and data quality in conjunction with the BI solution. The key issues facing the implementation team included •

Understanding in-depth analytics related to sales and customer data

•

Efficiently moving data into sales, customer, and financial data systems, as well as third-party systems

•

Improving audit controls around actual financial data

•

Improving data quality for better visibility into enterprisewide datasources

As a result of the BI system and the associated data integration and data quality solutions, the manufacturing company has realized more-efficient data integration, retrieval, and reporting. Specifically, the company can generate standardized and ad hoc reports, freeing employees to focus on data analysis rather than data retrieval. This has led to an increase in data accuracy and the ability to slice and analyze it in many different ways. The company also has the ability to budget and forecast based on business drivers such as staffing costs at the employee level or costs per unit or per employee. ORACLE’S DATA INTEGRATION STRATEGY Oracle Data Integration Suite provides a fully unified solution for building, deploying, and managing complex data warehouses or as part of data-centric architectures in an SOA or BI environment. In addition, it combines all the elements of data integration to ensure that information is timely, accurate, and consistent across complex systems.

Oracle Data Integration Suite provides a fully unified solution for building, deploying, and managing complex data warehouses or as part of data-centric architectures in an SOA or BI environment. In addition, it combines all the elements of data integration—data movement, data synchronization, data quality, data management, and data services—to ensure that information is timely, accurate, and consistent across complex systems. Oracle Data Integration Suite meets today’s enterprise data warehousing needs and addresses emerging trends in data services and federation, MDM, and real-time BI. It is a unique solution because it includes a powerful, native ELT approach. Most older solutions “push down” optimization, so data transformations still occur inside ETL engines and require the physical data to transit over the network and through their engines. However, Oracle’s native ELT solution doesn’t require separate hardware for the ETL engines. These ETL engines can be deployed directly on the target or even the source database. With a unique data integration platform that is architected for performance and productivity, Oracle Data Integration Suite provides a high degree of flexibility and modularity. It is a cornerstone for comprehensive data integration. By providing broad support for diverse IT environments, it executes on Oracle’s commitment to hot-pluggability. Regardless of the database or applications within your IT

State of the Data Integration Market 2008-2009 Page 27

ecosystem, Oracle Data Integration Suite can be optimized to drive the highestperformance bulk or real-time transformations. Oracle’s vision is to combine and enable these capabilities from within a next-generation, unbreakable SOA that will continue to drive business value within your enterprise for many years to come. For more information about Oracle’s data integration solution, please visit oracle.com/goto/ODI. CONCLUSION

Oracle’s State of the Data Integration Market Survey and analyst research have both shown that the market for data integration technologies is growing. While data integration is deeply rooted in data warehousing and ETL technologies, it is showing accelerating growth largely because it is a critical component for many technology imperatives. Clean, reliable data is essential for accurate BI; reusable data services are important for SOA; and data integration must be considered before developing an MDM strategy. In conjunction with these technologies, data integration offers businesses improved agility, better customer intimacy, and lower cost structures. Comprehensive data integration products include data movement, data synchronization, data quality, data management, and data services. Oracle’s Data Integration Suite provides all these capabilities in a single, unified solution that can deliver timely, accurate, and consistent information from multiple systems. However, companies need to consider the full scope and business objectives of their data integration needs before beginning any implementation. Careful planning and governance processes will increase the success of any initiative. Companies that follow the best practices outlined in this white paper—think big, start small, and act quickly—are most likely to realize the promise of data integration.

State of the Data Integration Market 2008-2009 Page 28

State of the Data Integration Market 2008 Authors: Dain Hansen, Jyothi Swaroop November 2008 Oracle Corporation World Headquarters 500 Oracle Parkway Redwood Shores, CA 94065 U.S.A. Worldwide Inquiries: Phone: +1.650.506.7000 Fax: +1.650.506.7200 oracle.com Copyright © 2008, Oracle and/or its affiliates. All rights reserved. This document is provided for information purposes only and the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any other warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose. We specifically disclaim any liability with respect to this document and no contractual obligations are formed either directly or indirectly by this document. This document may not be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without our prior written permission. Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners. 0408