Issue link: https://read.uberflip.com/i/1472351
MODERNIZING YOUR DATA INFRASTRUCTURE 4 IT imperatives for modernizing data infrastructure Technology most definitely does not stand still, which is why CIOs are under pressure to modernize their data infrastructures to support business-driven transformation. For many years, IT teams successfully built and main- tained large, general-purpose databases or data ware- houses to store a variety of information that different departments could access to run the business. However, with the massive amounts of data that organizations are creating and collecting, this one-size-fits-all strategy no longer works. "Given the explosion in data volumes and the types of data that customers are dealing with, organizations quickly reach a point where they need a different tech- nology," says Rahul Pathak, Vice President for Analytics at AWS. "With a general-purpose database, you'll end up in a dead end at some point because you're trying to do everything reasonably well, which means you can't be excellent at any one particular thing. A more focused approach lets you build something where there's no com- promise on performance, functionality, scale, or cost." Technologies such as machine learning and artificial intelligence (AI) are also driving the need to rethink data infrastructure. On-premises data warehouses cannot handle the processing power and storage required to run machine learning algorithms on petabytes of data. "While businesses have been using data analytics forever, the emerging areas of machine learning and AI are pro- viding capabilities never before achievable," says Gabriel. "CIOs can, and should, be leading that charge by under- standing how to work with the business to build those capabilities where they make business sense." CIOs also need to help their organizations better under- stand all of the data they are collecting, since it likely lives across cloud services, in on-premises servers and storage systems, and on employees' work devices. "Knowing the data you have available or could have available, and how it needs to be utilized to support an- alytics, is problematic," says Gabriel. "This has gotten far worse with the exponential increase in the data we can now capture and store relatively cheaply. The rise of SaaS availability has created muddy pools of data that exist within a company that are often unknown to IT or to the departments that could leverage that data." While the cloud has amplified some of these challenges, it's also a big part of the solution. A tightly integrated, cloud-based ecosystem of storage, compute, and ana- lytics provides better performance, increased scalability, and seamless access to data in ways that are unlikely or impossible with on-premises infrastructure. Benefits of data modernization in the cloud Modernizing your data infrastructure by migrating storage, data, and analytics services to the cloud can deliver many benefits. Here are a few examples: Operational and cost efficiencies Samsung's migration to a cloud-based rela- tional database helped reduce its monthly database costs by 44% and maintenance fees by 22%. Performance and availability Cathay Pacific saw a 20% performance improvement after moving its on-premises passenger revenue optimization system to the cloud. Scalability Moving from an on-premises data ware- house to a cloud-based data infrastructure allowed Nasdaq to increase transactions from 30 billion records a day to 70 billion records with no disruption. Security and compliance Since moving to the cloud, NuData's fraud- detection service has thwarted millions of attacks daily—protecting more than 100 million accounts every month, with 99% accuracy and a sub-0.1% false-positive rate. ML services on Amazon SageMaker have resulted in a 60-70% velocity increase in determining fraudulent activity.