By IT Brew Staff
less than 3 min read
Definition:
With the advent of the cloud, then AI, there’s more pressure than ever on data centers to deliver peak performance with as little latency and downtime as possible. In addition, massive datasets and AI models require data centers to modernize with the latest in storage, servers, and networking equipment. An aging data center, for example, might have severe issues with delivering the power, cooling, and security necessary for the dense GPU clusters powering AI workloads.
But modernization is also more than just updating the hardware in a centralized data center. With organizations’ data and cloud-based workloads often spread across multiple facilities, as well as on-premises servers and edge devices, modernization can often transform into a sprawling effort encompassing hardware, software, and infrastructure. As part of any modernizing effort, IT teams must assess the workloads and business goals supported by their data centers, then choose the right infrastructure—for instance, a company that wants to process data with extremely low latency might opt to shift most of their workloads to a constellation of edge devices and on-premises servers instead of reserving a set of racks in a remote data center.
IT pros tasked with modernization must juggle a number of challenges, including the energy constraints of the local power grid, the need to transition aging infrastructure to new with as little downtime as possible, and having the budget to purchase (and continually refresh) the latest and greatest servers and processors. With the demand for AI and data analytics continually rising, though, data centers have little choice but to modernize.
