The Apparent Connection between CIO's Success and Business Outcomes
To the Cloud & Beyond A Death Knell for Private Data Centers?
Mastering Partnership with a Remote Data Center
Understanding the Business First
Datascience: The Three Lessons Learnt
David Elges, Chief Information Officer, DC Government
Privacy and Ethics in Data Governance
Ken Knapton, Chief Information Officer, Progrexion
Fighting Fragmentation: How Capital Markets Firms Can Revamp Data...
Neha Singh, VP of Product Strategy and Innovation at Broadridge Capital Markets
Leveraging the Cloud for Data and Analytics
BRYAN GOODMAN, DIRECTOR, AI ADVANCEMENT CENTER AND CLOUD AT FORD MOTOR COMPANY
Thank you for Subscribing to CIO Applications Weekly Brief
How To Apply Machine Learning In Data Center
The digital twin brings all stakeholders together to plan and take control of their data canter’s performance and business impact.
Fremont, CA: Data centers operators can use Artificial Intelligence (AI) and Machine Learning (ML) technology to increase uptime, reduce energy consumption, promptly detect possible problems, and protect against cyber-attacks.
Major hyperscalers, for example, have built in-house AI to assist use cases like cooling. Smaller businesses, on the other hand, can benefit from AI/ML by utilizing AI-as-a-Service on cloud platforms.
By incorporating AI/ML technology in their products, data center companies are slowly making it easier for their customers to start using it. Specialized silicon, for example, is designed to execute complicated mathematical and computing tasks more efficiently. Because most AI use cases are currently fairly restricted, these AI chips can be trained for specific purposes like pattern recognition, natural language processing, robotics, network security, and automation.
AI is maturing, meaning that its abilities are constantly expanding as the cost curve falls. As a result of these two trends, data center operators will be able to include AI/ML in more of their products. For instance, RISC-V and other open-source technologies are decreasing the barriers to purpose-built "building blocks" that can focus on efficiency, performance, and scalability in ways that have never been possible before. Moreover, AI/ML can be applied to the data center's mechanical and electrical equipment to enable actionable insights and automation, saving money for the operator. This requires integrating traditional physics-based modeling approaches with state-of-the-art ML techniques using data from the Internet of Things (IoT) sensors. ML and physics-based modeling both have their strengths. Combining the two leverages the best of both worlds to solve complex data center issues involving mechanical and electrical equipment.
Data center design and administration should also incorporate digital twins so that the 3D virtual counterpart can imitate its physical behavior under any operational environment. In addition, it includes virtual representations of the facility's building parts, such as power, cooling, and IT system components from all major OEMs, as well as the whole data center ecosystem.
The digital twin brings all stakeholders together to plan and take control of one's data center's performance and business impact. All in one system, the digital twin enables enabling visibility to decrease operational risk, eliminate process bottlenecks, and enable the examination of "what-ifs."
Statistics center digital twins are physics-based, allowing them to mimic the performance of a new configuration rather than relying just on data. A detailed 3D representation of the data center space, architecture, mechanical and engineering systems, cooling, power connectivity, and the raised floor's weight-bearing capability are all included in a physics-based digital twin. This allows operators to predict, visualize, and quantify the impact of each data center change prior to deployment, giving them the confidence to make informed decisions.