The Modern World of Data Centers

Keith Engelbert, Chief Technology Officer,...

The Modern World of Data Centers

The Triple Bottom Line and Data Centers

Rob Nash-Boulden, Director, Data Centers, Black...

The Triple Bottom Line and Data Centers







4 Ways How AI Streamlines Data Center Management

Enterprise Technology Review | Wednesday, October 16, 2019

AI is transforming the firms' legacy approach towards data center management, raising potential ideas, and game-changing innovations.  

FREMONT, CA: A data center is an integral part of the business processes. Data centers are responsible for harnessing data, which can be used to spot new opportunities. Primarily, an efficient data center leads to smarter business moves, higher profits, more efficient operations, and happier customers. However, an exponential rise in data and power requirements are raising pressure on conventional data centers. Technology contributes to the advancement of data centers in the future. Technologies like artificial intelligence (AI), internet of things (IoT), augmented & virtual reality (AR & VR), and cloud computing, are raising pressure on the traditional enterprise infrastructure. The increased workload is further impacting the amount of power consumed by data centers with additional costs required for cooling purposes. The situation is expected to become worse in the future. Thus, organizations must come up with solutions that will manage complex data centers and fully secure them irrespective of the amount of data that needs to be dealt with. Here is how AI can contribute to addressing the requirements of modern data centers.    

AI: The Key Technology for an Efficient Data Center

The cost required to power a data center is estimated to be around 10 percent of the overall cost to maintain the data center. The power prices are estimated to account for 15 percent of the total expense of a data center by 2021.

Further, with an increase in data, significant resources and cost will be required to keeping the data centers cool. However, AI can address the above challenge as it can help in reducing the amount of energy required for cooling, irrespective of location. For instance, Google managed to offset the energy requirements of its data centers by 40 percent. Google accomplished the feat by using historical data that sensors had already collected and then incorporated AI to train neural networks that comprised the AI system. Eventually, the systems could then predict how various actions would affect energy consumption at a particular time. The predictions were fed back to the data center’s control system, which incorporated the required actions accordingly.

 AI Analysis for Effectively Utilizing the Data Centers

It is essential to ensure that the infrastructure of the data center is being utilized properly. Incorporating AI for the analysis of data captured from IoT sensors can allow the firms to utilize their data centers effectively. AI analyses will be the key to provide the information needed to maintain the system at 100 percent availability while significantly offsetting the cost. Thus, AI analysis will lead to predictive, instead of scheduled maintenance, eliminating the requirement for the servers to be offline for a long time. 

 AI-Enabled Digital Twins

Developing and maintaining a digital twin is another idea that allows machines to have a digital form as well as a physical one. Such facility aids in the planning of machine placement and use. A digital replica allows factories to experiment on virtual platforms before physically trying those experiments on the physical systems. 

 In the case of business information modeling (BIM), a construction approach enables the creation of a detailed virtual walk-through of entire buildings. AI algorithms actively engage in the development of a digital twin, planning, or BIM automation to constantly search for the optimized version and implement it accordingly. It can either mean altering the physical stack or the development of detailed plans in case the physical movement or building of data centers is required.

 AI Powering the Decentralized Data Centers

With an increase in the creation and demands of data, organizations have realized that centralized management of data is both time-consuming and costs them significantly in terms of speed and agility. Edge computing is aimed at countering this issue. Typically, the technology aims to bring the processing capabilities closer to the source of action (or the network edge), such as a bank branch or a retail outlet. AI will assist the organizations in implementing, managing, and optimizing several mini data centers across their decentralized networks. AI will empower the decentralized units by allowing them to process the relevant data, identify actions in a particular scenario, and implement them accordingly.

 Thus AI offers a plethora of opportunities to the firms to improve data center capabilities. It is upon the organization to realize the applications of AI that will best suit their data center requirements and take the right course of action.