Data architecture

Data architecture

14-Jul-2021 05:52:47 am

Data architecture is the process of standardizing how organizations collect, store, transform, distribute, and use data. The goal is to deliver relevant data to people who need it, when they need it, and help them make sense of it.
Traditionally, if a business strategist wanted data, they requested IT, which would then develop a system to deliver it. However, this lengthy, labor-intensive process had the unfortunate effect of delivering something other than what the strategist wanted. Availability and growth of real-time data from internal and external sources have led business strategists to demand more and faster insight from data. 
Modern data architecture promises to bring business strategists and technical expertise together through a well-designed process. Together, they can determine what data is needed to propel the business forward, how that data can be sourced, and how it can be distributed to provide actionable information for decision-makers.
Modern data architecture is expected to be innovative and inspiring; agile and responsive; and compliant and secure, Fifty percent of organizations lack sufficient AI and data literacy skills to achieve maximum business value.
According to Emma McGrattan, at Data Summit Connect 2021 the following are the key five tenets of a successful enterprise data architecture:


#1 Encompasses all data types and sources
Getting access to all data types, including structured, semi-structured, and unstructured data, and all sources across channels is critically important for many company initiatives such as Customer 360, compliance with GDPR and other regulations, KYC protocols, and others.


#2 Elasticity
Cloud provides clear advantages in elasticity noting that businesses want to be able scale up and down their usage to address as needs and demands change and some want this done automatically, while others want an administrator to approve new capacity.


#3 Delivers value on day-one
Today, many organizations believe a platform should start delivering value on day one, and they should be able to measure value and see that value in the first couple of days. The tolerance for long-running projects is disappearing. Organizations also want an environment that meets their business needs, and to not have to pay for something that's going to accommodate their rare peak workloads all the time. Some companies are choosing to have a tiered data strategy, for example, in which they keep their corporate crown jewels on a database appliance but move their second-tier applications to the cloud to free up valuable space rather than purchasing more appliances which will take time to become fully useful


#5 Resiliency
As you build out a data platform, you've got to make sure that you don't have any single points of failure, said McGrattan. It has to be built to be resilient and with layers of redundancy so that you can guarantee your 24x7 business operations. And if a component were to fail for whatever reason, there must be automatic failover to a backup so that it the process is invisible to users.
The modern data platform is designed to serve a range of purposes, from securing sensitive corporate data to answering questions from non-technical users