| | FEBRUARY 20208 IN MY OPINIONBy Raghu Radhakrishnan, CEO & MD, TmaxSoft India 019 is behind us, and you know what that means. It's time for technology predictions for 2020. And there are plenty of those to go around. If you search for them on Google and look at the summaries, you'll notice that almost all of them mention AI and robotics. You can't read an article about 2020 without seeing one or the other or both.So, to make things interesting, this article is bucking the trend and focusing on other technology, leaving AI and the robots to Gartner and Forbes (except for one small exception at the end). So, without further ado, here are our top 5 predictions for 2020. 1. It all begins with GPUGPUs (graphics processing units) have been driving advances in games and professional graphics for more than a decade. But lately, data scientists and IT departments have discovered that GPUs are good for a lot more than a realistic fight in Super Smash Brothers. GPUs are now providing computing power for big data analytics. An example is using massive amounts of images of malignant and benign tumors to train models to show when a tumor needs investigating. The latest development in this space are GPU databases that are more flexible in processing many different types of data, or much larger amounts of data. Because big data and real-time analytics are needed now more than ever to deliver customer experiences that drive competitive advantage, GPU databases will be a hot item in 2020.2. Say hello to MDBSMDBS is the abbreviation of multi-database system, sometimes also called a multi-model database system. Companies today can have different models of databases, often residing in different business units and storing different kinds of data and transactions. Accessing data from those databases outside the business units has been tricky at best. Resolving the issue has usually required moving the data somewhere else using ETL (extract, transform, load) processes that are time-consuming and compute intensive.MDBSs offer access to the data stored in these various individual, separate databases. They can query data using different models including relational, NoSQL, and more. As the number of cloud databases and NoSQL databases mingling with relational databases increases in 2020, MDBSs will be just what the doctor ordered to enable access to the different data models.3. Hybrid IT with a purposeHybrid cars. Hybrid mattresses. Hybrid clouds. Hybrid fireplaces. The world is full of hybrids -- including hybrid IT. Hybrid IT is a mixture of on-premises data centers, private clouds, and public clouds. For years, hybrid has been organic. Enterprises have added to cloud to their IT, but most still have on-premises systems, too. After cloud proved to be a successful infrastructure for Amazon, Google, and other digital companies, the push from analysts was to move everything to the cloud as soon as possible. However, for most companies, that is not practical, cost-effective, or secure.So, as this decade of the cloud is coming to an end, enterprises have realized that hybrid IT doesn't have to 2Devise Thoughtful Hybrid IT Strategies! Raghu Radhakrishnan, CEO & MD
< Page 7 | Page 9 >