“As you store, so you retrieve” – architecture that delivers

The primary purpose of storing data is for its retrieval at a time of need in a manner that best serves the underlying purpose.  Transactional databases were founded upon a “row-based approach”.  However, complex analytics operates on vertical “column” based subsets of data that matter.  Storing the information using a column based approach would ensure that the data can be readily retrieved at decision time.  Hence, data must be stored in a manner conducive to its retrieval.

Big Data retains the Competitive Edge

This is one of the primary messages that I took away after reading (and listening) to the interview of Barry Ralston, Assistant VP of Data Management at Infinity Insurance by Dana Gardner. Infinity, a billion-dollar property and casualty company, works with large volumes of data about customers, policies, payments, and etc.  Ralston asserts that Big Data is a strategic weapon for Infinity to retain its competitive edge.

Here are other key points that surfaced from this interview along with some of my own thoughts.

1. Analyze data that matters. Ralston explains that the traditional concept retrieving whole rows of data elements is no longer an effective approach for complex analytics. When customers purchase policies, Infinity might be more interested in the top ten policies that are in demand rather than other details. Therefore, it makes more sense to retrieve, sort and analyze only the data elements that matter.

2. It is all in the architectureColumn Storage.  Warehouse management is a well-honed science now, where pallets and items are placed in a manner that they can be easily accessed and retrieved. The storage architecture significantly impacts the ease of retrieval. Infinity’s solution follows this simple yet powerful principle.

3. Improve performance at the origin. Time is vital when it comes to process data into information. Performance is a key architectural tenet that must be applied at the source of the data element. This will obviate the need for downstream reactive measures like in-memory databases and additional hardware.

4. Time-to-decision matters. Are we there yet when navigating the Big Data traffic jam? Tools exist today that can process data to deliver information at decision time.

5. Get ahead of the game. “Insurance is an interesting business, in that as my product and pricing people look for the next great indicator of risk, we essentially get to ride that wave of competitive edge for as long a period of time as it takes for us to report that new rate to a state,” says Ralston. Predictive analytics are vital to look ahead at what is coming to retain the competitive edge.

To retain a competitive edge in their industry sector, enterprises must proactively apply complex analytical techniques on data that matters. And they must do this quickly. This will protect them against unforeseen market downturns, such as consumers insuring themselves against unexpected calamities.

So, what is at the heart of this solution? Vertica Analytics Platform.

Wondering what the letters V E R T I C A stands for? Here is what I think.

Want to know more about this tool? HP’s Big Conference for your Big Data.

Want to connect with me? Twitter, Facebook, Linkedin and Journey Blog.