Informatica Corp. is the overall leader among data virtualization software vendors, followed by IBM, Composite Software Inc. and Denodo Technologies Inc., according to a new report from Cambridge, Mass.-based IT analyst firm Forrester Research Inc.
The new Forrester Wave report—which evaluates and ranks data virtualization software vendors based on criteria such as overall strategy, functionality and market presence— found that Informatica, IBM, Composite, and Denodo each offer a strong combination of features such as support for real time integration, data services modeling and caching. The four data virtualization “leaders” are followed next in the ranking by SAP AG, Microsoft, Oracle Corp., Stone Bond Technologies, L.P., and Red Hat Inc.
“SAP, Microsoft, Oracle, Stone Bond Technologies, and Red Hat are Strong Performers [and] each offers a viable option to support particular use cases,” the report reads. “Red Hat continues as the most substantial vendor supporting an open source data virtualization solution.”
Data virtualization tools add an abstraction layer to IT architectures that enables information from multiple data sources to be integrated in real time, near-real time or batch, as needed. The technology is used to support many types of applications such as real-time business intelligence (BI) reporting, enterprise search and high-performance transaction processing.
Organizations typically run data warehouses to aggregate cleansed and highly structured data about transactions that can later be analyzed for business insights. A key benefit of data virtualization is that it uses standard interfaces to provide access to data wherever it resides, regardless of whether that information is structured. For example, an organization could conceivably use data virtualization to combine information from an enterprise resource planning application, a Microsoft Excel file, a data feed from a social media site, and a data warehouse.
“Customers know they want business data,” said Noel Yuhanna, a principal data management analyst with Forrester and co-author of the report. “It doesn’t really matter whether it’s structured, unstructured or semi-structured. At the end of the day, it’s all business data.”
A data virtualization market breakdown
Informatica made several strides in data virtualization in late 2009 when it unveiled Informatica Data Services 9, which included advanced features like “on-the-fly” data quality and profiling and the ability to integrate cloud applications, according to Forrester.
Forrester found that IBM’s data virtualization portfolio is buoyed considerably by its services organization, which helps companies complete and maintain data virtualization initiatives. IBM’s primary data virtualization offering, IBM InfoSphere Information Server, integrates with several complementary products including IBM’s Information Analyzer, QualityStage, DataStage and Change Data Capture and Federation Server products.
Composite has improved its offering in recent years by adding advanced data virtualization capabilities like support for unstructured data sources and cloud integration. Composite, which “supports some of the most complex data virtualization deployments in existence,” also offers a lower cost of ownership than larger players in the market, according to Forrester.
Forrester gave Denodo Software props for simplicity and ease of use. The analyst firm said Denodo is also particularly good at integrating unstructured information with structured information.
SAP is in the process of extending its data virtualization offering to support highly complex deployments and real-time integration across applications, databases, unstructured data and external sources, according to the report. Microsoft, meanwhile, offers strong support for devoted users of the Microsoft platform.
“Microsoft has enhanced BizTalk Server, SQL Server Integration Services, SQL Server, and Visual Studio to deliver a framework to support data virtualization on Windows,” the report reads. “It continues to extend this framework to support cloud, unstructured, and semi-structured data and to improve its performance, integration, and security.”
While Oracle offers a strong set of technologies that can be used for data virtualization, the company is not currently marketing a packaged offering, Forrester reports.
“If Oracle customers want to see more support from Oracle for data virtualization—tell your account executive,” the report reads. “Forrester expects Oracle to return to the data virtualization market in the coming years, as it grows too large to ignore.”
Stone Bond’s main product, Enterprise Enabler, applies a metadata-driven approach to discovering and integrating data sources. Forrester reports that Stone Bond is well suited to specific use cases, such as the integration of data from medical devices with billing systems and cloud applications.
Red Hat’s open source approach gives users flexibility when it comes to integrating unique data sets or platforms, according to Forrester.
“Although not all of the Red Hat components are open source, its data services platform offers a viable option to support unique and complex integration requirements,” the report reads. “Today, financial services, government agencies, retail, insurance, and telecom firms are using Red Hat to support data federation, with real-time integration for service-oriented architecture-based applications.”
BI tools help complete the puzzle
Data virtualization packages are good at providing a means to access information from various sources. But they’re not as heavily focused on helping users gain valuable business insights from that information, according to Yuhanna.
That’s why organizations mulling a data virtualization project should look for options that play well with existing BI tools, and organizations without a BI environment should consider investing in one. Yuhanna said some of the more popular BI tools include Oracle Hyperion, SAP BusinessObjects and IBM Cognos.
“Data virtualization vendors do have basic dashboards, but I don’t think that’s sufficient to deal with all the requirements,” he said. "You really have to look at third party [vendors] who specialize in BI.”