Information governance is not just about the data or the technology that is used to manage data, it is about an entire ecosystem within an organization – including people, process, and policies.
And thus, creating and implementing an effective information governance strategy is a shared responsibility, said an expert during an event recently.
The Ministry of Transport and Communications (MoTC) recently organized a webinar on ‘DataOps in the Era of Big Data & AI’. Diaa Eldin Ali, technology expert from IBM, highlighted how data is no longer considered the by product of business activities. He shared the importance of having information governance as a strategy in the planning of the data landscape.
The session emphasized information governance as an important area which many people are missing on during their strategy planning for their data landscape, especially Small and Medium-size Enterprises (SMEs).
The objective of the session was to give more practical overview of the role of information governance in obtaining successful Data & AI practice in business. The event discussed data operations and how to manage data operations quickly and focused on organized data which according to the speaker is the most time-consuming and most comprehensive area that takes most of the time during software development.
He said, in order to have well organized data you need to have three things which are the pillars of DataOps. Firstly, to know the data you are working on to make the project successful, secondly to know the data is trustworthy through data quality checks and curation and thirdly how to use this data.
He said, “Knowing your data is very important capability as data discovery and classification is number one thing required to know that the data is governed and secure. Because not knowing the location of data is a big problem when auditing the records.
So, the discovery of where the sensitive data is and classifying the sensitivity whether it is public, confidential or internal is vital”. “Trusting the data is a data quality concept tied up with curation of the data. Usually, data quality is a big concern because not all data is high quality. To use and understand the data is to define the concept of data dictionary or a business glossary. Data dictionary is considered as an index of definitions of all the data landscape within the organization.
There are many organizations without data dictionary or glossary which leads to a lot of misunderstanding and miscommunication within the internal teams within the organization. And all this has a common infrastructure of data management and under all these layers data software can be deployed by any cloud or in Microsoft, Amazon, Google,” he added.
“The different capabilities that come under DataOps is important to have proper data governance in any organization. The three main pillars under DataOps are – the data governance pace which is about understanding and governing the data, the data quality pace, and data consumption pace.”
He said, “Today organizing data or governance of data is renamed to DataOps to fit into modernization of the software. The main purpose of this is speed of applications because in the era of modernized software there is no room to have software development for two three years as it was 20 years ago.
Today every application has a new version every month, sometimes every couple of weeks, the reason being easy to develop and fast to produce new versions and the same is going on in the software industry for the enterprise applications for different industries.”
DataOps is an agile methodology for developing and deploying data-intensive applications, including data science and machine learning.
