MarkLogic‘s Joe Pasqua dropped by to talk about the challenges some companies have with islands of data that are hard to integrate. This, he points out, makes it difficult to bring their data together allowing them to use Big Data analysis techniques to ferret out insights and better serve their customers. He suggests that a No-SQL database, such as MarkLogic. No surprise there.
Why a No-SQL database?
Many enterprise databases are constructed using the rigid relational model. Some older databases are likely to be built using CODASYL network data structures. Still others might be using special-purpose multi-way binary tree structures. Modern Big Data applications are being built using a broad array of new approaches to store machine data, documents, and other forms of non-structured data. Getting access to all of this data creates a dizzying number of challenges for enterprises including the following:
Data fields are likely to have been specified differently for each type of data store. Unifying this so that the data can be stored and accessed in a unified fashion can be quite challenging.
Enterprises don’t have the time to stop production so that data can be unified. This means they must come up with a way to keep data in its original form for production applications and still be able to access it in a unified way.
Enterprise applications need high levels of availability, reliability and must offer ACID compliance to rule out problems caused by distributed access and multiple people/programs reading, adding and deleting data.
Many of today’s databases keep data, meta data and indices in one place making it difficult to use modern distributed processing techniques to scale up and scale out.
Once the data has been put in a single place, it can be challenging to find a way to access it in a way that is useful for analysts, developers and executives.
MarkLogic believes it has the answer
Pasqua then went on to point out that MarkLogic has been offering a solution has been in use since 2001 in data centers of over 550 customers that addresses all of these issues. He pointed out that MarkLogic’s database offers a flexible data model that makes it easy to integrate data from multiple sources, a very powerful search and query function that makes it easy to find and use needed data items, support for a semantic data model making it easy for business analysts to access information without having to learn a programming language, and is ACID compliant to assure enterprises that their data is safe.
Dan’s Take – An interesting take on making data useful and available
Although our conversation was rather short, I heard all of the right things. Pasqua’s discussion of MarkLogic’s flexibility, performance, scalability, ease of use of a clever distributed data model made me want to know more. I suspect that if you speak with them, you’ll come to the same conclusion.
I hope to speak with one of MarkLogic’s customers soon and will publish a “Making the IT Decision” eBook based upon that conversation.