The most important component for enterprises of every scale is building secure, state of the art and resilient data centres which maintain high levels of security. In a post-covid world where everything that one does has been shifting to a virtual platform, safety is at risk and companies have to scale up and boost their data centres to prevent data attacks from taking place.
“The key processing hubs behind the global rise of cloud computing data consume huge amounts of energy. From 2010 we had only 2 Zettabytes of data and today we have, in 2020, close to 59 Zettabytes of data. By 2024 we will have 149 Zettabytes of data,” Binu Chacko, Partner-Cloud Transformation and Cybersecurity Leader, EY said. These hubs are necessary as they manage and maintain data that powers the daily lives of people all over the world. Cyber attacks and security breaches are other concerns that plague facilities and enterprises. One of the main questions that arise is an ideal way of designing large scale data centres that do not compromise their operation and reliability.
Various applications which were traditionally used within the organisational premises are now moving beyond the boundaries, people are taking charge from their homes and are being productive in every field. This is leading to a change in the movement of data centres- from structured to unstructured. “That’s an area that I am particularly very interested in” exclaimed Vamsi Ponnekanti, Head of the Technical Sales, India SAARC, McAfee. “It’s (The pandemic) made the organisation boundariless and perimeter-less,” he said.
Face-to-face conversations have turned into virtual meetings through video calling applications, homes have turned into smart homes, cars turned to computers on wheels, and more. Another question that takes precedence is the way organisations have been catering to this data trend explosion and the changes that require coping with the growth.
Digital transactions are now the new normal. Transactions, ranging from a mere Rs 5 to upwards of crores are taking place every minute. So, building systems flexible and agile for these transactions has become an essential part of the data centres. One of the highlights of this post-pandemic world is that there’s a visible improvement in the modernization of infrastructure and applications.
Building efficient data centers
Enterprises are encouraged to have more secure and modern facilities for maximum efficiency. In brief, the four main pillars revolving around the evolution of the data centres consist of automation, software-defined networks, hybrid adoption methods, and monitoring. “Manageability was the biggest problem organisations faced. How am I going to manage those islands of data with one control pane” Anantharaman Balakrishnan, Vice President and Managing Director, Nutanix, said.
The underlying part of every infrastructure is still the same. It still consists of CPUs, servers, Operating systems, and other traditional means but now has been repackaged as virtualization and hypervisors. “The difficult part is modernising the applications. The easy part, comparatively easier part is to modernise the infrastructure” said Vinit Sinha, Director- Cybersecurity, Mastercard.
The harder part is the modernization of applications, making them microservice-based, controlling them and driving the needs on the infrastructure. This is a long term process where companies can take anywhere from 4 to 7 years to implement the same. However, the more critical issue is testing. Secure and high bandwidth is necessary for testing these pieces before going live for the customers. A great customer experience is the main driving point for every enterprise. Breaking the monolithic applications into microservices is a big challenge, along with testing and the cost involved.
Security and testing play a vital role in moving applications from an on-prem structure to a cloud-based structure. While these data are put onto the cloud, enterprises are at the risk of direct attacks. “While you’ve put some of these pieces on to clouds you’re under the direct attack vector,” Sinha said.
Earlier companies and data centres had the traditional methods of security which consisted of firewalls, routers, and other protective measures but now due to these data being on the clouding platform, at any moment can there arise a situation of a data leak, which is why security and testing are essential. Contactless applications are more on the focus now. Even if the progression rate is slow, enterprises are witnessing an improved efficiency and cost-management. Organisations need to start adopting and implementing the concept of shared responsibility. “Organisations have to start adopting some of these pieces into their culture,” Sinha added.
Shared responsibility means segregating responsibilities—giving a particular role to a particular person or sector, giving the vendor a role, network providers a role, application designers a role, and creating a robust framework to manage all these responsibilities together. Configuration end of the data is the responsibility of customers. “Beaches in industries happen due to misconfiguration and weak credentials” he said.
Data Localization Requirements
Handling sensitive data’s complexity has increased exponentially. Different regions and countries have their own set of laws and regulations to comply with and follow. Even after creating a framework to manage these data, one may not be able to effectively implement these due to the changes in these laws. “Even if you’re creating a framework to really manage some of these pieces, you may not be able to do it very effectively,” said Sinha.
It is necessary to create frameworks for specific regions to ensure those data localization methods work in accordance with the laws of that particular region. “Making sure you’re continuously in sync and rhythm with how new applications, new infrastructure is being used,” said Ponnekanti. “If I had to cause harm to you, I’d go after that application. I’d have to attack your infrastructure,” he said.
Data localization has not changed for traditional enterprises and big companies. In India, for instance, data centres are not affected in the big enterprises, but more to the start-ups. Companies should make sure that a copy of their data is stored on the cloud for security purposes. “Acceptance towards agility is now in hours and minutes,” Singh said. “We still don’t allow any internet, directly internet access once you’re on the corporate network,” he said. Privacy is a bigger question and requires thorough and heavy regulations. The number of attacks has increased significantly and platforms require some basic guidelines to manage some or most of these attacks.
