Dr. Harrick Vin, Head, Systems Research Lab, TCS Innovation Labs – TRDDC

Over the past decade, we have witnessed the emergence of a plethora of technologies for simplifying and transforming enterprise IT. Yet, most enterprises are struggling to design and operate IT environments that cost less while ensuring adaptability and supporting business agility.

This challenge occurs because of two reasons. First, the complexity and scale of modern enterprise IT environments makes it difficult to obtain a holistic understanding of the as-is state of an enterprise. In most cases, IT environment designers and operators only have a silo-based understanding, along dimensions such as application, infrastructure, operations, and technology stack, among others, which makes it very difficult to predict the impact of a change, leading to resistance for introducing change. Second, each transformation technology has costs, benefits, risks and side effects; further, their applicability is governed by several technology and business constraints. Because of this, one size does not fit all; what works for one enterprise often does not work for another. Hence, a custom transformation strategy is needed for each enterprise.

Derivation of such a custom strategy requires an enterprise to answer three key questions:

(1) The what: What should the target state be?

(2) The when: When should one migrate from the as-is state to the target state?

(3) The why: Why do we embark on this transformation program – what is the business case?

Today, most enterprises rely on manual processes, intuition, and experience to derive a simplification and transformation strategy.Such an approach not only takes time, but often also results in many implementation surprises leading to many failed transformation programs. To overcome these challenges, as part of the simplification theme, we have been developing an analytics-led approach to simplify and transform enterprise IT. This approach relies upon facts (data collected from an operational environment) rather than intuition and experience, and systematizes and automates the process of deriving a custom transformation strategy for each enterprise.

During 2009-10, we began development of a suite of enterprise IT transformation tools that embody this philosophy. The tool suite includes:

eModel: A tool for modeling and describing enterprise IT environments

ePrep: An enterprise-IT-aware tool to extract-transformload data from an operational IT environment into the enterprise data models

eAnalyze: A collection of as-is state analysis tools (e.g.,the iPredict tool for performance and reliability assessment, and the iSupport tool for analyzing production support activities)

ePlan: A collection of tools for automating the process ofderiving a custom transformation strategy (e.g., the iTransform tool for transforming IT infrastructure)

eExecute: A tool for systematizing the execution of the proposed transformation strategy

eCollab: A tool for supporting collaboration, knowledge management, and workflow across teams driving such end-to-end enterprise transformation programs Development of the suite is a large, multi-year effort that involves close collaboration with several COIN™ partners. In particular, our work leverages emerging technologies and tools from strategic alliances (e.g., Cisco, BMC, EMC and NetApp) and startups (e.g. ServiceMesh). Further, we collaborate on some long-term research issues with several academic institutions (e.g., IITBombay and IISc-Bangalore in India, and University of Massachusetts at Amherst and the University of Texas at Austin in the USA). Some of our early results and success stories have also been presented at international conferences.

More Information

Virtualization Monitor (joint R&D with Hitachi)
Managed evolution (SURe)
Enterprise Cloud Computing
Algorithms and Optimizations
Next-gen Infrastructure
External projects - ITER, Telescope