Building Crisis-Proof IT Infrastructure for the Enterprise

Over two months into the coronavirus lockdown, we’re starting to see a clear divide between prepared and unprepared businesses. Some had the necessary infrastructure to weather COVID-19 lockdown, while others have had to play catch up – to their major disadvantage.

It is with this in mind that we brought Commvault and the Future as a Service together for an in-depth look at how to create crisis-ready IT infrastructure. Don Foster, VP Storage Solutions at Commvault, and Neil Cattermull, Director at the Future as a Service, joined us to share their insights and how organisations are coping with the data tsunami and keeping the show on the road during a crisis.

A lot of content online right now suggests that there has been a data explosion since the pandemic hit. But for Neil Cattermull, this is a little bit of a fallacy. To start our webinar on crisis-ready infrastructure, Neil looked at the data statistics from before the pandemic and showed how data is undergoing a meteoric rise – and has been doing so for years.

• In 2019, there more than 4.5 billion people online
• Netflix’s content volume in 2019 was more than the entire US TV industry in 2005
• 188 million emails were sent every minute in 2019
• At the beginning of 2020, the digital universe was estimated to consist of 44 zetabytes (44 trillion gigabytes) of data – 40 times more than the whole amount of stars we can see in the visible universe!

Clearly, data has been increasing rapidly, and would doubtless continue to do so. But that’s when the unthinkable happened – the pandemic came and countries were forced into lockdown to contain its spread. Whole industries shut their doors overnight and a substantial amount of the world’s economy ground to a halt. However, contrary to what’s happening in the wider economy, data saw an explosion as more people moved online.

Countless organisations were unprepared for this. They had to transition quickly to a remote working approach, but the swift gear change put untold pressure on cloud service providers. We all know the stories, Neil said, so he wasn’t going to name and shame those that struggled to scale at the start of the crisis. Thankfully, these providers have now caught up – but the pandemic has already done untold damage to organisations.

It was at this point that Neil highlighted a perhaps controversial point. We’re all familiar with the phrase, “Data is the new oil”. But for Neil, this is a categorical error. Data is not the new oil. Oil is a finite resource that you use and consume and then it’s gone. Data couldn’t be more different. It is its own resource, with data begetting more data and organisations building up their databases to generate new value.

Neil showed that data is self-perpetuating, creating its own value in a way that oil can never do. But this belief in data as the new oil may have contributed to the warped view that influenced many IT budgets pre-pandemic. If data is something to use up, you don’t need to invest in the infrastructure to handle it. With 5G, AI and other emergent technologies reshaping the world does business, a data tsunami is coming, and only the organisations with the IT infrastructure to handle it will prosper.

So for Neil, it’s organisations who adopt a multi-cloud approach that will thrive most. He discussed the possible factors that could transform what the future looks like, citing both pessimistic and optimistic forecasts. But regardless of what tomorrow looks like, Neil was clear: only a multi-cloud hybrid approach will deliver success. With Don, Neil discussed how organisations will also utilise containerisation in particular to handle the data tsunami. More data is coming, and if you’re not agile or ready for anything, you can and will be in huge danger.

Neil then handed over to Don, who had been contributing his own input throughout Neil’s half of the webinar. To find out more about what they discussed, including their views on containerisation, download the full webinar now.

Don focused on the infrastructure IT leaders need to maintain operations, even in times of crisis. For Don, there’s always going to be something that could cause a challenge or crisis, whether it’s a natural disaster or human error. You therefore need to be agile and ready for anything.

What does that mean in practice? Is it a case of turning to the next emergent technology? Making a big investment and establishing a whole new IT infrastructure?

Instead of turning to some new technology, there could be another way. Instead of innovation, we should look towards standardisation to protect our organisations in times of crisis. Don argued that no one put this better than Henry King in Fast Company.

“At a time when we are constantly being told to value the new and the different, it may come as a surprise to learn that the standard, the shared and the common can be strong drivers of transformation.”

It isn’t the case that value is only in the new and different. The standardisation of processes can be better. With a core set of fundamentals to define your infrastructure and organisation, you can maintain consistency so you don’t have the mass fragmentation of IT infrastructure that makes it difficult to weather a crisis. After all, with mass fragmentation it is harder to reorganise efforts in a crisis.

In addition, new trends are emerging and creating new types of challenges:

• Multi-cloud adoption is increasing, with 84% of organisations having a multi-cloud strategy
• Software-defined storage deployments are accelerating, and by 2024 50% of global storage capacity deployed SDS
• Global organisations are quickly moving to containerized applications, and by 2022 more than 75% of organisations will be running containerised applications in production

For Don, there’s one answer to these challenges: software-defined storage.

With software-defined storage, organisations can establish the scalability, predictability and agility they need to weather a crisis like the Covid-19 pandemic – and more. Don and Neil talked at length about the new infrastructure requirements IT teams will need, including ensuring it embraces a self-service, platform-integrated, hybrid-ready and policy-based approach. Then, Don went on to reveal the traits your infrastructure must have to deliver in a time of crisis. Automation, service-based economics and policy-driven, multi-tenant software is a necessity, and you can only deliver these if you embrace abstraction, establish resiliency, and implement scalability.

For Don, this means standardising your IT infrastructure. Standardisation, abstraction and resiliency provide choice, and Commvault’s Hedvig solution provides that. This is a software-defined storage platform that can span multiple on-premises and disparate cloud environments to seamlessly provide multi-site resiliency. Availability is built into the platform and realisable on a per volume granularity basis.

In particular, Hedvig is multi-cloud inside. It was built from the ground up with experience gathered from inventing systems like Amazon Dynamo and Apache Cassandra, both of which have deployed in large internet estates. What’s more, typically native storage primitives in the cloud don’t have cross-zone/cross-region fault tolerance built in; but Hedvig does. It can sit between the application and the storage primitives in the cloud to provide such capabilities seamlessly for greater application availability and reliability. Find out more about Hedvig here.

Finally, Don shared an example of how Commvault has helped to provide crisis-proof infrastructure through Hedvig. He used the case of a current production cluster set up for an American university. It is a 3-site HA cluster with two data centres being close by (with sub ms ping latency) and the third data centre being about 170 miles away with a 5ms ping latency. Hedvig acts as the underlying storage platform across all three data centres.

All data is replicated across the three data centres, and due to Hedvig’s quorum semantics the applications don’t suffer the long latency to the third datacentre. Any writes that might have failed are tacked and fixed by a background task that Commvault calls re-replication.

If any site fails, the applications seamlessly migrate to one of the remaining sites, and all I/O operations continue without any interruption. Since all previous writes have been committed to at least one of the remaining two data centres, RPO is basically zero so this is truly HA. There is no downtime, at least from a storage point of view, even on a complete site failure.

In a nutshell, software-defined storage is the critical tool that enables organisations to store and run applications beyond the boundaries of its data centres – and establish a truly crisis-proof IT infrastructure.

To learn more about Hedvig and the other insights Don and Neil had to share on the day, as well as to hear their answers to our audience’s burning questions, access the on-demand webinar now.

This is just a brief summary of everything that our esteemed speakers addressed during our webinar on crisis-proof IT infrastructure. If you’d like to hear more insights from Don Foster and Neil Cattermull, you can access the on-demand webinar for free.

Alternatively, if you’d like to learn more about Hedvig, we can help. Get in touch at [email protected] to learn how Commvault can help your organisation to establish an IT infrastructure that is ready for anything.

Ready to learn more about how technology is evolving in a post-Covid-19 world? Register for any of our upcoming webinars to help plot your organisation’s future.