In-Depth
ETI Does Hassle-Free Data Integration
ETI touts is own unique spin on data integration: precompiled data integration executables. It's one of several key differentiators for the veteran competitor.
Data integration pure-play vendors have fallen like dominoes -- albeit at rather protracted intervals. There was Acta Technology (which was acquired by Business Objects), DataJunction (acquired by Pervasive Software), and Ascential Software (acquired by IBM Corp.), Sunopsis (acquired by Oracle Corp.), DataMirror (acquired by IBM), and Business Objects SA (acquired by SAP AG), among others.
What's surprising is that there are still so many data integration players left, and that the data integration market continues to thrive.
One data integration veteran may be worth a second look. Evolutionary Technology International (ETI) was founded 18 years ago, and unlike many of its data integration competitors, ETI markets its own (homegrown) legacy connectivity solutions.
Getting at mainframe data, after all, is ETI's bread-and-butter business: its ETI Solution V6 boasts connectors to a wide range of data sources, including legacy platforms of all kinds (e.g., MVS and VSE on the mainframe side, OS/400 and VMS in the minicomputer segment), Unix, Windows, Linux, and all major relational database management systems.
ETI Solution V6 bundles data profiling, data cleansing, and data monitoring facilities, along with change-data-capture capabilities.
ETI's special sauce, says Wyatt Ciesielka, vice-president of North American sales, is its code-generation engine, which produces extracted, cleansed, and transformed data in the form of an executable compiled in a variety of languages -- including C, Java, and COBOL.
Code generation isn't as much of a throwback as you'd think, says Philip Russom, senior manager for research with TDWI. ETI's approach is popular with government agencies, especially with the Department of Defense, as well as large banking and financial institutions, because of its inherent security. After all, compiled data, checksum and all, is auditable and governable data.
"Code generation is one of [ETI's] key differentiators," Russom says. "With code generation, you can go to unusual platforms, [including] platforms people have forgotten about, like the HP-3000."
Notwithstanding its desirability in security-conscious environments, ETI's code-generation capability has other benefits. "You might want to preprocess data on the mainframe to reduce its volume," Russom explains. "If the mainframe is a source of lots of data for your warehouse, you need something like ETI that will preprocess data."
In Big Iron environments, it boils down to an issue of cost. In spite of IBM Corp.'s efforts to make the mainframe more affordable, mainframe processing cycles are still somewhat pricey. "Typically, what happens on the mainframe is that you rarely get to run data against a mainframe database. Instead, what usually happens is that there's a routine that some guy wrote back in 1988 and it dumps more data than you need into a big flat file, and the flat file somehow finds its way on to an open system. Many companies would like to process the flat file on the mainframe, [but] they only need 10-20 percent [of the data in the file]."
Make no mistake: ETI isn't a mainframe-only proposition. Its customers use it to connect to all major operating system platforms.
Nevertheless, its mainframe-connectivity feature, like its code-generation capability, is a key differentiator, says Ciesielka, especially in a data integration market where so many players treat the mainframe -- home to between half and three-quarters of mission-critical data -- as a second-class citizen.
In this respect, ETI has an especially good story to tell, Ciesielka argues: its integration products can shift data bi-directionally within z/OS itself, between z/OS and other open systems (e.g., Windows, Linux, Unix) -- and between z/OS and other so-called "legacy" platforms (e.g., MVS, VSE, OS/400 and VMS).
Elsewhere on the differentiation front, ETI also markets a Built to Order (BTO) integration capability -- basically, canned, fully documented integration workflows (including all requisite cleansings or transformations) for mainframe, distributed, and proprietary applications -- which promises what amounts to turnkey connectivity, according to Ciesielka.
"This [BTO] is so much faster than using something like JDBC or ODBC," he notes. "There's no software licensing, so it's cheaper than other [platform] solutions. It's much easier than hand-coding, it especially reduces processing time for large data loads, [and] it generates a full audit trail."