Image courtesy of Shutterstock
Industry Leader Q&A with Attunity’s Lawrence Schwartz
Attunity recently added new capabilities to its solution suite with the acquisition of BIReady's data warehouse automation technology, which eliminates the complex, manual tasks of preparing data for BI and big data analytics. Lawrence Schwartz, Attunity’s vice president of marketing, spoke with DBTA about BIReady and other Attunity solutions for customers dealing with big data projects.
DBTA: How does the addition of BIReady change Attunity’s position in the market?
Lawrence Schwartz: BIReady is a company that focuses on getting data ready for analytics. Attunity is very focused on how customers get the data from where it originates to where it needs to go. We have been successful with the strategy but when data is being moved into a data warehouse there are often a number of extra steps that have to occur in order to prepare the data in advance - and that is a process that often takes a lot of manual effort: You have to design the data warehouse, the data marts, load the data, transform the data. That can be a process that takes weeks and months. We have been in the business of simplifying and automating data movement and, by being able to automate that one step further so that we automate the schemas, automate the transformation, and automate the history, maintenance and logging, we have an end-to-end capability. This fits very well into our overall business model and that is the value that BIReady brings to the business.
DBTA: Is this a requirement that emerged in customer deployments?
LS: One of the things customers want to do is roll out new services and offerings quickly and they want to take advantage of all the latest big data technologies very easily and so they have come to us a lot in the past to help them do that. This would come up in conversations as an area that they were struggling with. Traditional ETL (extract, transform, and load) vendors have always been very focused on the E and the L and achieving high performance there, and on replication and what that delivers in real time environments. This allows us to also deliver that T that also needs to be done from time to time at the data warehouse level.
DBTA: What are the types of data that your customers are working with?
LS: Often, data is going from the transactional model, the OLTP, and then they want to pull that data in and do advanced analytics on it. To get the performance out of it they have to normalize the data and build the correct business model to set up the data warehouse. So, the data is coming from traditional sources - CRM, ERP, and other transactional systems. It doesn’t mean that the technology can’t do more but that is what a lot of the customers are working with because they need to have that full ACID compliance on the transactional level and that is something for which they still rely on a structured database.
DBTA: What are the kinds of opportunities that customers are seeking to take advantage of?
LS: Customers want to make their processes more effective and efficient. Say, for example, an insurance company is trying to pull together its financial reports for doing analytics. Without the right tools and without BIReady, it would take it a few weeks to get that done. Once you automate that process and wring out the complexity, that report can be done in two days - so now all of a sudden a monthly financial report is much more current, the company has a much better ability to pull analysis out of it, and it can make adjustments more rapidly to the business. This allows organizations to spend more time on doing the actual analytics.
DBTA: When you talk about big data, you are mainly referring to volume issues, not variety?
LS: I am taking about the large data warehouses, and all the systems that our customers traditionally work in right that they want to do more with - from Teradata, Netezza, Oracle, and Microsoft. There are also a number of solutions that we have developed in the Hadoop space and as time goes on and as we integrate all of our solutions there is certainly opportunity to go ahead and look at ways to bring that all together. However, BIReady solves a key part of the automation process that people are grappling with now. It goes that extra mile into the data warehouse beyond what they had been able to do.
DBTA: What are some of the key changes that you have made to help customers in the Hadoop space?
LS: In Q4 2014 at the Strata conference, for example, we announced Replicate 4.0, with Hadoop support, to help reduce the complexity of moving big data to and from Hadoop. That has enabled us to work with a couple of different scenarios. People want to use Hadoop for data lakes - as big repositories for long-term analytics. They want to put data in those formats from traditional data sources. It enables people to do a lot of preprocessing in Hadoop and then put it into more capable data warehouse that has much faster processing.
It allows customers to think about extract, transform, and load and how they can do that differently. Now, in Hadoop, you can do the transformations, use extract capacity and power that you have on the nodes to do the transformation process there and then leverage a lot of the Attunity simplicity, ease of use, and fast performance on the extract and load side in conjunction with that. This has opened us up to many more types of scenarios, and even though we just offered it starting in Q4, we already have some major wins with customers that will leverage it.
DBTA: What is the most common misunderstanding that comes up with customers around big data?