Big Data Needs a Big Engine


Integrate Fast Transforms, Improve Application Data Security

Optimize Application Performance


IRI software's data manipulation engines are fast, scalable, and designed to accelerate third-party applications. IRI libraries and executables are easy to integrate and inexpensive to license. Make C, C++, Java, COBOL, or system calls to do one or all of the following in your software:

  • High volume data transformation (sort, join, aggregate, pivot)
  • Reporting (batch, delta, detail and summary BI formats)
  • Protection (field-level encryption, masking, de-ID)
  • Conversion (of both file formats and data types)


In the new IRI Voracity platform for big data management, you can also run the same data transformation, protection, and reporting jobs in Hadoop MapReduce, Spark, Storm, and Tez seamlessly from the IRI Workbench GUI.


If you call the thread-safe sortcl_routine() library in IRI CoSort, your applications can do all of the above, and in one I/O pass. From sorting to summing, and reporting to encryption, you can quickly power and secure your applications; and they process data with thread-safe IRI code.

Mask PII in Your Applications, Comply with Data Privacy Laws

If you develop or test applications that require access to production data, you must mask it in ways that comply with business and legal rules. IRI FieldShield software masks personally identifiable information (PII) in your databases and flat files so you can protect personal data privacy while still retaining data realism.

Test Applications with Safe, Intelligent Test Data

Use IRI RowGen if you develop applications and need to stress them with the right range of accurate, relational test data. RowGen does NOT require access to production data to create realistic, referentially correct test data. Populate any database or flat file format, plus structured reports. Even assess the portability of your applications across platforms since RowGen can also manipulate the data it generates!

Use the Fastest Sort Engine

Every computer science curriculum covers sort routines, so you know what they are. Traditional sorting operations are non-linear; as the data doubles, the process of putting that data in order triples. So as your volumes increase, what ran quickly last month now seems to take longer and longer to finish. If production windows are starting to close, you will want to consider an embedded sort alternative.

IRI CoSort has been consistently reported to be the fastest and most scalable sort engine for big data on large Unix and Windows servers. CoSort has been successfully integrated and made available for wider distribution by AMS, CA, Cincom, Experian, Fiserv, Kalido, Mereo, Sabre, SAS, SPSS, and Unisys, to name a few. Learn how affordable CoSort is to integrate.