Pervasive messaging

Written By:
Content Copyright © 2010 Bloor. All Rights Reserved.
Also posted on: The IM Blog

A question I am frequently asked about some companies with a diverse portfolio of products is, “what exactly do they do?” A case in point is Pervasive Software, which does not find it easy to describe itself. Go to its website and you will find the following text: “Pervasive Software helps companies get the most out of their data investments through agile, embeddable software including on-premises and cloud-based services for data management, data integration, B2B exchange and analytics.” Hardly pithy.

Go a step further and you will find that the company identifies five product categories:

  • Pervasive integration (which, though it is not mentioned up-front, includes the company’s data quality offerings);
  • Pervasive PSQL, which is what is still probably better known as Btrieve and is the company’s embedded database;
  • Pervasive DataSolutions, which provide pre-packaged connections to things like and QuickBooks, which most people would think of as a subset of data integration but which users might not, since they are primarily business people;
  • Pervasive Business XChange, which provides Web-based business-to-business data interchange which, again, most people would categorise as a form of data integration;
  • Pervasive DataRush, which is a (very) high performance engine, SDK and library of operators for supporting functions such as high-volume data preparation or in-flight analytics.

So, what have we really got here? A long-lived database with a substantial partner community and a variety of products that are essentially about moving data. If we take the database first you might think that the product was mature, and indeed it is, and you might also think that it wasn’t going anywhere. But there you would be mistaken. For example, in the latest release, PSQL v11, the company has introduced multi-core optimisation in a similar fashion to the way that technology is used with DataRush. This is important because your average application will run about 20% slower across multiple cores than it will across one. This is partly because of lower clock speeds, partly because of syncing requirements across the cores and partly because you have multiple caches to consider. Therefore Pervasive has optimised its database to make use of multi-core parallelism in a way that is invisible to existing applications but which will provide a 300 to 400% performance improvement. That’s pretty impressive: I’m surprised that other database vendors aren’t doing it.

We should also mention Pervasive DataCloud. All Pervasive DataSolutions and many custom services created by the company’s partners, reside on Pervasive DataCloud, which lives in AWS’ EC2 environment and enables Pervasive and other cloud developers to create and provision on-demand data services.

As for moving data, that’s not all that Pervasive does with it; it’s also about doing stuff with the data during the process of movement. And by doing stuff we don’t just mean the sort of transformations that you would expect from a data integration product (which, of course, Pervasive does) but doing real work. For example, the latest release of DataRush enables the application of data mining principles as the data flows through the DataRush engine to, say, a back-end data warehouse. We might call this operational analytics.

Thus, aside from the database, Pervasive is about managing and manipulating data in motion or, if you prefer, in-flight. Of course, that’s not all it does, but it seems to me that the company could tell a clearer story based around this concept, with a coherent message about what Pervasive does as a company that makes it unique and how that uniqueness will help it to solve its customer’s business problems.