“On its own, Big Data means nothing. It’s got to exist, coexist, and work with the other data,” said Ron Huizenga, Senior Product Manager at IDERA Software. “The good news in the industry is [that] we’re past Big Data being the shiny ball that people are chasing,” and now we’re trying to assess what Big Data means to the organization and how to “knit it into the fabric” of the enterprise environment, for both IT and business users, he said. Huizenga and Joy Ruff, IDERA’s Product Marketing Manager, discussed the use of Data Modeling and other tools for collaborative Data Governance with DATAVERSITY® at the Enterprise Data World 2017 Conference.
About IDERA
IDERA has been in business since 2003, evolving from its original focus of SQL Server Database tools, to now being “a broader Database Life Cycle Management company,” Ruff said.
“The portfolio ranges from basic diagnostic management of the database, through compliance management, to backups and performance needs. There’s a whole host of products now that address the database life cycle from start to finish,” she said.
IDERA has recently expanded their multi-platform capabilities with the acquisition of Embarcadero Technologies, “so it’s not just SQL Server anymore.”
With their March 2017 release, IDERA also integrates with Atlassian’s JIRA issue tracking product, so team members can edit tasks and user stories in the ER/Studio Team Server collaboration platform and view related model changes. According to IDERA’s website, these updated Data Modeling and Data Architecture tools feature enhanced Data Governance capabilities, making it possible to close gaps between Data Modeling and development, and to improve collaboration between data professionals and business users.
“Businesses are now recognizing how important data is. They’re also recognizing how complex the data landscape is where we have NoSQL data products,” said Huizenga. Understanding the entire data landscape in the organization is not just about data itself – it’s enabling the business to tie together Enterprise Architecture, Data Architecture and, “folding in the business processes that reference that data and everything else,” he said.
“With the increasing use of Agile and high-velocity development methodologies, Data Modeling must be incorporated into the workflow to ensure data quality and consistency.”
“In the past, a lot of organizations or teams used Agile as an excuse not to create full deliverables,” often resulting in data modelers reverse engineering and documenting databases after the fact, he said. ER/Studio is “geared to be able to handle the high velocity of Agile,” he commented.
“So, that means we’re equally as good at being able to handle the more traditional methods. You’re still tying into your requirements. Your cycle’s a little bit different, but all the same capabilities are there.”
Projects can get moving without being completely documented because, “It doesn’t matter if you do that upfront or if you do it a little bit later in the cycle. As long as you’re making sure it’s one of the criteria to complete the process,” he said.
Without Data Modeling, Big Data = Big Mess
“It’s funny because people talk about Big Data and it’s like, ‘We’ve got to capture this Big Data and we’ve got to dump it into our Data Lake,'” he said. “But, it’s like an episode of ‘Hoarders.’ What are you going to do with it? That’s the real thing you’ve got to look at. It’s even more important now to be able to distinguish and figure out which part of that data is important to me.”
Yet finding meaning from Big Data can be overwhelming. “If you ever look at one of those data streams, it’s like drinking from a fire hose and trying to pick up the drops of water that you need to correlate to pull together. Big Data’s the same problem,” he said. It’s important to have a visual model, so that both IT and business people “can truly understand that data’s journey through their organization, what’s been happening to that data, and which business processes those changes have happened within.”
Ruff adds, “you need to be keeping it up to date. You’re not just creating this data model once and then putting it in an archive and never looking at it again,” which is why IDERA’s ER/Studio has Change Management built in, she said. “We’re encouraging the use of the universal mappings and the Enterprise Data Dictionary. This stuff has to live and evolve as the business grows.”
Huizenga said:
“When people think about ER/Studio, they first think about the Data Modeling, but it’s a lot broader in terms of overall Enterprise Data Architecture. We also have Enterprise Architecture capabilities, and in particular, business process modeling capabilities as well.”
“We can do integrative process modeling that references the data source from your data models. We do data lineage, not only just to source the target mapping from when you’re building out your Data Warehouses,” he remarked. “We can interrogate ETL tools and actually bring that back and create the data lineage diagrams, stitching the data landscape together.”
Huizinga used customer data as an example, saying that the typical company has customer data “scattered across multiple databases,” with different owners. “We’re able to reverse engineer and map those databases [using] a concept called Universal Mappings, where we can link those concepts together through our repositories,” allowing a business to see all instances of ‘Customer’ in the organization, as well as which databases it’s deployed in. “So, it really allows you to tie that overall picture together. That modeling is just the start of a framework for Data Governance overall,” he said.
“If you look at any data model, it’s not just a picture: It’s a statement of business rules,” said Huizenga. “So, if you look at any entities, the relationship between those entities, that relationship itself has a business meaning as to why it exists, and that’s what we’re trying to do is make sure that we’re capturing that in the metadata.”
True Collaboration
Huizenga says it’s now more important to emphasize the concept of ‘business-driven,’ compared to ‘data-driven’ Data Architecture. “We really need to recognize the fact that the business owns the data. And again, they own the business glossaries. They define what that data is.” IT serves as custodians of the data, he said. “They make the decisions with the data.” The entire organization should be equipped to understand what that data is.
“Whether it’s in their operational systems, or whether they’re actually using it in traditional BI Data Warehouse systems, or in the Data Lake, and they’re trying to do Data Science and Data Analytics on it to try to extract out more information,” he said.
Huizinga says that ER/Studio also allows business and technical users to work together: “We tie in business glossaries with the visibility and the movement of data back and forth, or the metadata back and forth between the glossaries and the modeling environment,” so it’s visible to the entire organization. He stressed the importance of knowing not just what has been changed, but why it was changed, and what the requirement was that necessitated the change.
“Anybody who has sat across from an auditor [who is] saying, ‘Why did you change that database?’ knows what we’re talking about here. It really is a full collaborative platform for enterprise Data Governance as well.”
Metadata extensions can be created in the Enterprise Data Dictionary and then can be referenced in all models. Setting up classifications for Master Data Management in ER/Studio integrates them automatically, he said, “So you can say, ‘Show me this classification. Where’s my Master Data?’ and it can actually show you where all of it is, and what it’s called.”
Data Governance is important to every organization, he said, whether it’s regulatory approaches or privacy regulations that are coming in.
“We’ve always had the capability [to create] security classifications for your data. We have some that are right out of the box that we can generate, or you can build your own – and those are visible right on your diagrams and all the published metadata.”
“For the evolution, we continue to see the growing need for DBAs, developers, and data modelers and architects to understand each other and how they work together,” said Ruff. The goal is to be able to ‘connect the dots’ within that Enterprise Architecture so databases can run more efficiently and “tie back to the data models that define what the business is actually doing with their data. “And that consistency can run throughout the company,” she said. “Know what your data means and where it comes from, and how you’re going to use it to make your business successful. Novel concept, right?”
Photo Credits: ra2studio/Shutterstock.com