Infrastructure automation has already made an impact on data center productivity and efficiency. Now, machine learning and artificial intelligence (AI) are adding a new dimension that promises to improve these factors even further.
The aim is to create a self-learning environment in the data center that will enable automation tools to recognize problems and initiate responses without human intervention. Commentators are describing the process as an evolution of the software-defined infrastructure (SDI) into an artificial intelligence-defined infrastructure (ADI) that will support dynamic, real-time monitoring and management of the data center environment.
Learn and Respond
Artificial intelligence and machine learning are finding application in many industries. Forrester Research believes that investment in these technologies will increase rapidly as businesses recognize the vital role they can play in automating the process of identifying problems and providing solutions.
Machines are capable of doing this, with or without the aid of human intervention, making the process of responding and providing solutions much faster. Machine learning systems progressively improve using historical and current service and technical data.
Another task that benefits from machine learning techniques is the evaluation of massive complex data sets. Many businesses have archives of valuable service information stored in formats that are complex and too large for IT teams to make sense of.
Machines can learn the skills to determine useful patterns in huge and complicated data sets. This is a promising area for development where a combination of programs that acquire knowledge with massive storage capacity and ever increasing processing speeds opens massive new opportunities.
Applications in the Data Center
Machine learning can automate monitoring, maintenance and provisioning functions carried out by IT. This is important because of the growing complexity of the data center environment where staff must track and respond to thousands or even millions of events.
Automated systems based on machine learning can initiate remedial action. The systems learn from incidents so they can solve problems based on evolving knowledge. By analyzing incidents and changes, these systems help create a real-time environment that can maximize performance and minimize downtime. Machine learning can automate monitoring and maintenance, freeing the IT team for strategic tasks.
From SDI to ADI
Deploying machine learning is the next step in the evolution of software-defined infrastructure. It adds intelligence and learning to static source codes or scripts to create a dynamic self-learning/self-healing environment known as artificial intelligence-defined infrastructure (ADI). Commentators believe this can transform infrastructure from an under-utilized capital asset to an efficient operational resource.
An ADI can support dynamic provisioning of infrastructure from a pool of resources based on its ability to understand and learn from patterns of consumption, configurations, dependencies and usage patterns. It can also contribute to right-sizing infrastructure based on patterns of usage and consumption of resources to deliver more efficient resource allocation.
An Intelligent Solution
Commentators believe that the future of enterprise infrastructure is intelligence and self-management. This will not only improve infrastructure efficiency, it will free IT staff from routine monitoring and maintenance tasks. IT is then in a position to add more value to the business by focusing on new service developments and other transformation initiatives.