Cloud: The problem of seamless migration

As enterprises migrate to public cloud, they face challenges as well as new dimensions of work management. Varoon Rajani, Co-Founder & CEO, Blazeclan Technologies, interacted with us to share his thoughts.

What are the typical problems that large enterprises in India face while trying to embrace digital transformation? What is the typical migration journey?

At Blazeclan, we believe in cloud-led and cloud-enabled transformation. Any business looking to go digital or scale in today’s fast paced environment has to have cloud as its core of their technology landscape. Since our inception, we have been a user of AWS cloud for all the projects that we have delivered for our customers. We have helped companies to move entire data centres to cloud across APAC.

The most common problem faced by large enterprises is how to move infra seamlessly and predictably to cloud. Governance and security on cloud is another major concern. Optimising applications and expenditure on cloud is one more area where CIOs look for our expertise. Building skill sets on cloud tends to get difficult and dealing with the changes that cloud brings in terms of processes and people can be quite challenging. Companies take different ways to migrate to cloud. This is primarily driven by the reason to move to cloud. We have seen 2 typical patterns:

1. First one is to save cost. This is usually done by shutting down data centres. To do so, the primary step is to calculate the TCO (Total Cost of Ownership) and ROI (Return on Investment) of moving to cloud. This will also include the Public Cloud platform selection. Once the ROI and TCO are clear and approved, enterprises typically migrate a few servers as proof of concept that helps them understand the technical as well as governance challenges of migration. Based on the success of the pilot migration, a time-bound mass migration plan from the data centre to the cloud is created. The migration journey to save cost is often driven by hardware refresh or data centre contract renewal

2. Second pattern is driven by innovation and automation. This is for those enterprises where focus is to improve the speed of innovation and response to customer demands as more and more companies are becoming digital. For this, the approach is to migrate application by application. Here, typically the companies look at refactoring or modernising the applications while migrating the cloud. This approach allows the companies to use the benefits of cloud like elasticity, scalability and fault-tolerance by upgrading their applications. This a longer approach, but one that suits well for customers who are undergoing digital transformation. The cost benefi¬ts with this approach may be a bit delayed.

You have a diverse range of clients like finance and media companies along with the BSE. Is it a one size fits all or are the needs and implementations of these entities totally different?

We help our customers with cloud led and cloud enabled digital transformation. We believe, an organisation can be truly digital only if they are agile and nimble in responding to their customer and market demands. To be truly agile and nimble, organisations need agile infrastructure along with agile processes, which is what public cloud is. We help our customers with cloud advisory and migration services, automation of their entire IT tool chain along with managed services of their cloud infrastructure. There is no one size fi¬ts all approach for cloud adoption. Each company is driven by their own business priorities. As explained earlier, two of the major priorities that customers have is rapid digitisation or cost savings. Each company makes choices based on these priorities.

With microservice architectures becoming more prevalent and cloud automation enabling the use of APIs seamlessly, customers will now have access to applications with best of technologies from across different clouds. Though this may take a longer time to become a reality in larger enterprises due to their limited skill sets and capabilities, but we see a trend of using a service from the cloud based on the value it provides.

What problems will the Indian industry face while moving from a scale-based model to a skill-based one? What are the factors that individual companies need to take note of?

The way the technology ecosystem has evolved, applications today are tightly integrated with the infrastructure. Previously there was a clear segregation between the teams that developed the application and the team that procured and managed the infrastructure. Due to this, there were silos in the operations and in the skill sets. People were labelled as infrastructure specialists or application developers, with knowledge of one or more programming language.

With Cloud computing, infrastructure can be managed by code. This means that developers can manage infrastructure to an extent and hence need to understand and learn infrastructure management, networking, security in a better way. For infrastructure specialists, there is a lot of automation and optimisation that cloud brings in, which really frees up a lot of their time. With these changes, companies will have to look at ways to deploy large teams of engineers they have in different areas effectively by cross-skilling them. Cross skilling people will reduce team sizes and time of operations as automation will be the fore front. However, there will also be companies who would be looking at cross skilling from a rapid growth point of view and hire more such people to scale in the industry quickly

The cloud can provide a lot of agility and scalability to all manner of individual companies. Can you explain this in greater detail?

Public cloud infrastructure is designed in such a way that it is highly scalable and is on demand. What this means is that anyone looking to buy any infrastructure, for example a server to host an application, can do this within a few minutes with a few clicks. Compare this to the earlier scenario for enterprises where there was a longer procurement cycle to purchase a server, harden it and then deploy the application.

As public cloud is on-demand and pay as you go, the smaller businesses or start-ups do not have to invest heavily on IT infrastructure. They now have access to the same scale and quality of infrastructure as any of the larger enterprises. This was not possible earlier since setting up IT infrastructure involved a lot of cost. This has given a level playing fi¬eld to smaller businesses who are now able to scale faster at a lower cost and are competing aggressively with their larger competitors. You can see the number of start-ups have increased in the last decade, a lot of it is possible because of the access to on-demand, pay as you go IT infrastructure.

With cloud computing 2.0 closing in and XaaS becoming a reality, what can the industry expect in the next 5 years?

While a lot of the services are available as services, we are still far away from having a world where XaaS is a reality. More than XaaS, what we see customers bene¬fiting the most is from the on-demand availability of services, pegged with pay as you use pricing. Customers do not want to invest heavily in products upfront, and then worry about utilization. They would prefer to use on demand services as and when required. Majority of the start-ups and more and more enterprises are now moving to this model.

AI, Machine Learning, Big Data and mobility are fast changing fields. How is the new cloud technology integrating all these changes?

Cloud computing has reduced the costs of storage signi¬ficantly and has increase the processing speed of data tremendously because of its on-demand and scalable infrastructure. Companies are now storing every single byte of information that is generated since the costs are very low. With this scalable infrastructure, companies are now able to spinoff large compute clusters to process the data and then shut it off without investing in setting up the infrastructure. Both these points have led to an increased adaption of Big Data, Machine Learning and AI within companies. Each major public cloud provider is investing heavily on the data processing, ML and AI capabilities and we will see a lot of innovation coming through over the next few years in this area. Without public cloud computing, very few companies would have been able to invest in Big Data, ML or AI.

What more can the Indian government do in promoting cloud computing in all the diverse sectors like healthcare, agriculture, education, land management and utilities?

If the government wants its Digital India initiative to succeed, it must whole heartedly adopt public cloud for its own projects. They should also encourage start-ups to venture into areas of e-governance, payments, education, agriculture among others to truly digitise these industries. For digitisation, cloud computing must be the core. The fi¬rst thing the government needs to do is to have a clear data residency regulation/policy in place along with a set of security compliance for the usage of public cloud computing in public sector. This will benefi¬t a lot of start-ups and companies to innovate without restrictions and at the same time use the best in class technology. Apart from this, there should also be awareness created among government departments at both federal and state level about the benefits of public cloud.