Cloud Hosting

By- Amith Singhee At the dawn of the Information Age in the 1970s, the role of Information Technology (IT) was limited to computing plumbing - to keep the networks and computers working. In the 90s and 2000s, it evolved into an enterprise shared servicesmodel that was essential for operational efficiency, cost takeout and decision support.

Today, IT is witnessing another shift that increasingly requires the Chief Information Officer organization to act as a partner in defining business strategy and driving topline growth via IT-driven business transformation. To realize this, the IT delivery platform that includes infrastructure, applications, processes and roles of people -needs to be scalable and adaptable tokeep pace with the rapidly changing business and operational needs, and, hence, transform to a hybrid cloud IT architecture.

The transformation will involve four phases: Advice for Cloud, Move to Cloud, Build for Cloud and Manage on Cloud. Artificial Intelligence (AI) will play a fundamental role across all these phases. To understand this better, consider three illustrative scenarios from the Move, Build and Manage phases that are foundational to hybrid cloud adoption across industries: application modernization, DevSecOps and incident analysis and remediation.

Modernizing applications for the hybrid cloud

However, the application today does not allow for an API-based integration and is architected as a large monolith that would not efficiently scale to support dynamic loads on the system.Since enterprises have many such monolithic applications, they would like to modernize these applications to a scalable and modularized architecture, leveraging microservices and cloud technologies like Red Hat OpenShift, SaaS and IaaS, deployed across a hybrid cloud footprint. However, such modernization can take substantial manual effort, which can easily span over one to two years.

By using AI, the retailer can substantially reduce the time and effort of modernization and achieve better outcomes. Here is how Beginning with the Discover and translate stage of the modernization process, AI can analyse all application artefacts such as source code, logs and architecture documents toestimate a topological model of the application. Next, algorithmic and AI approaches can be used to do a Biz-Ops Analysis andoverlay business and operational KPIs on the components of the topological model. Finally, at the Optimize and Re-engineer stage, AI and optimizationcan be used to generate a Biz-Ops optimized target design including architecture, refactored source code, DevOps configuration and deployment specification. The architectcan guide and orchestrate all the stages of this process in an interactive manner.

DevSecOps in a Hybrid Cloud environment

DevSecOps integrates development, IT security and IT operations in a unified approach, where high levels of automation shorten application development lifecycle, enable continuous delivery, remove agility barriers from security gates, and ensure high software quality.AI can be of tremendous value in enabling effective DevSecOps. During development, for instance, AI-enhanced static code analysis will detect security vulnerabilities and non-compliance issues in the development phase itself, as plugins within the developers favourite Integrated Development Environment. This will substantially reduce the delays that occur today from post-development security checks.

AIs role in incident analysis and remediation

In the operations domain, consider the situation when an incident has occurred, for example, a user webpage may be unresponsive, or a web service API returns a 502 error and that has triggered a ticket or alert. If a site reliability engineer (SRE) is tasked with resolving the incident, AI can assist with incident analysis and predict next best actions, understand causality to estimate root causes and recommend remediation approaches.

This is how it works - AnAI system, trained from a corpus of historical incident data, can analyse all available structured and unstructured data related to the active incident, extract important incident features in the context of deployed application topology and diagnose the issue better.

Secondly, a variety of AI techniques can be used to predict the next best actions that the SRE can leverage to further diagnose the incident in collaboration with relevant subject matter experts. Updates to the active incident data can then iteratively be used to further advance the incident analysis, diagnosis and recommendations towards resolution of the incident.

While these three scenarios are widely applicable to enterprises across industries, similarly significant application of AI is likely to be seen to most other parts of the IT delivery platform and application lifecycle.

The shift to experiential age

The dramatic improvements in AI over recent years have been driven heavily by use cases and datasets from the consumer world images, speech and text. With the emerging synergy between hybrid cloud and AI, we will witness tremendous innovation and business value in the enterprise IT world. Today, customers increasingly expect instant and seamless multi-channel engagement and data privacy, andenterprises need to meet or exceed these rapidly evolving customer expectations.

As businesses leverage technology, data and AI at an increasing scale and new business models rapidly emerge,they need toundergo business transformation at tremendous speed to stay competitive while continuously reducing costs. To conclude, AI and hybrid cloud are catalysing the journey of digital transformation by enabling a fundamental shift in the role IT plays in the business.

The author is Senior Manager Hybrid Cloud, IBM Research India.

See original here:
Galvanizing the new age of IT with AI and hybrid cloud - ETCIO.com