As part of Information Age’s Cloud and Backup Month, we are providing three CTO guides over the coming weeks on cloud and backup: cloud migration best practice, how to manage cloud infrastructure and cloud predictions.
The first guide focused on cloud migration tips. Organisations are migrating to the cloud for digital transformation (or digitisation) efforts — they are doing this to keep pace with growing customer demands and an increasing need to scale services.
One of the key lessons from the first guide surrounded automation. “Automate as much as possible when in the cloud; try to abstract the provisioning and use agnostic tools so that you can move workloads between cloud providers until you get a feel for which provider is best for that workload,” said Neil Briscoe, CTO at Cloud Gateway.
This guide will focus on best practice tips for managing cloud infrastructure post-migration, with insights from CTOs and cloud experts.
Use a provider
Walid Negm, CTO at Aricent, says that most of it should be done by the provider, “such as the delivery of integrated tools for monitoring application health.”
Adopting the cloud: Applications, pitfalls and choice
“Automate everything, for example, Infrastructure as Code, deployment and operational (high availability, scalability, disaster recovery etc). Re-architect applications that enable self-healing and resiliency. Ensure that the monitoring and instrumentation is also done as a declarative model and that the developer has control of this.”
Find a partner
Stephan Fabel, Director of Product Management at Canonical – the company behind Ubuntu, concurs and suggests that organisations need to find a partner to help with complexity and skills shortages.
“Deploying cloud platforms and services from multiple vendors can be seen as a rather complex process, especially when it comes to efficiently weaving multiple different clouds together in a way that doesn’t hinder productivity or innovation. The inherent technical challenges of managing several different platforms in a multi-cloud deployment — such as understanding exactly where data resides and having to integrate legacy systems with new platforms — has the potential to cause difficulties. They can also amplify already-existing business pressures, most notably responding to customer demands for service and innovation. This is especially true for those organisations that either lack the in-house skills, or fail to find a partner that has the capabilities required to manage such a transformation.”
Open source in enterprise: Necessary for digital transformation
“This is something many businesses have quickly realised as they have continued along their deployment journeys. Indeed, issues around technical and skills challenges were identified as critical learnings from managing multiple clouds by more than half (57%) of senior IT leaders.”
“The skills gap has been an extremely prevalent issue in the cloud world for some time, with many businesses either lacking the budget to meet the substantial salaries that people with cloud skill sets now command, or simply unable to find people with the required level of technical expertise. This highlights the importance of finding the right partners so that businesses can hand off the most complicated jobs to a team of experts.”
“However, it also highlights the need for better tooling for lifecycle management and operations. Lowering the barrier to entry, a solid choice of orchestration and management frameworks will take a pragmatic view on what’s needed to increase productivity around the day to day operations, and exceed expectations around even complicated processes such as upgrades of complex infrastructure software. This in turn will increase the pool of candidates businesses can pull from when looking to staff (or skill) up.”
Adopt adequate tools
Luis Weir, CTO for the Oracle DU in Capgemini UK, states that the cloud won’t manage itself.
How can you design, deploy and manage your APIs?
“Adequate tools have to be adopted in order to consistently manage all elements of cloud infrastructure, from IaaS to PaaS and even SaaS. However, the future being multi-cloud, identifying tools that can be used to manage cloud infrastructure across different cloud vendors becomes extremely important. So my tip would be take your time evaluating which is the right tool considering that most likely you won’t be putting all your eggs in a single cloud-vendor basket.”
Security is paramount
Aron Brand, CTO at CTERA Networks, says security is paramount.
“It’s important to remember that cloud infrastructure is inherently a shared service that resides outside your firewall, and we live in an era of daily reports of data breaches and other data loss events. If you’re a security conscious organisation and want to embrace cloud technology while minimising risk, I have three pieces of advice for you:
• Deploy all sensitive services within your firewall – either on-prem in a private cloud OR leveraging virtual private cloud technology from any of the leading cloud infrastructure vendors (AWS, Azure, Google Cloud, IBM Cloud, et al).
• Next, ensure that all data sent to the cloud is encrypted. Just as you wouldn’t access sensitive information over public wi-fi without a VPN, you shouldn’t use public cloud storage without proper encryption. Ensure high levels of encryption at-rest (AES-256) and in-transit (TLS 1.2).
• Last, generate and own your encryption keys, and don’t rely on a third-party. Doing so ensures that even if your data becomes exposed on public cloud infrastructure that it will be useless gibberish to any non-authorised user.
Edge to cloud: digital transformation and the shifting data landscape
Change: The new normal
Adam Evans, Director of Professional Services at Rackspace, advises that organisation’s, post-migration, build a target operating model that assumes change is the new normal.
“Assume that you will constantly be consuming new technologies, then build better processes to assess and bring them in effectively. This will help you deliver greater business value from these investments and avoid overburdening yourself with managing the growing cloud sprawl.”
A pipeline suitable for enterprise
Neil Briscoe, CTO at Cloud Gateway, provides a number of tips for managing cloud infrastructure:
• Automate as much as possible!
• Create and govern design templates – and stick to them!
• Have the network/security team manage the networking parts (routing, firewalls, subnets) and give the devOps teams a “secure play pen” to work within. Separate out the duties based on skill-set.
• Have thoughtful monitoring and alerting. You need a healthy balance between “noise” and needing to take action.
• Create a pipeline that is suitable for your enterprise and the way you and your teams want to work – don’t feel the pressure to go with what the crowd tell you to use.
The application portfolio
Chip Childers, CTO at The Cloud Foundry, offers some cloud management advice regarding an organisation’s application portfolio, and how you rationalise it post-migration.
“In one bucket you have the things that you don’t need to change because you have no demand to change it — you just cost optimise and that’s all you’re going to do. So, you can lift and shift if it’s valuable or just keep it in the data centre if moving it is too risky, just leave it alone.”
What’s causing the move to the cloud? It’s all about digital transformation
“On the opposite side, you’ve got new applications that need to be built because you have a demand from your business partners or the market overall, you need to build new software. That’s where you want to adopt cloud native software design practices, that’s where you need to really just focus on being good at software delivery. And then we’ve got the stuff in the middle.”
“The stuff in the middle tends to be the software that’s been around for a while but is super hard to change and you’ve got a lot of probably frustrated business partners that have a lot of pent-up demand that they want to work with you on. This is where you really need to carefully find ways for each application: to take it apart, maybe wrap it in an API, do something that’s going to free you up to respond more quickly; like you can with the newer style application and encapsulate risk as much as possible.”
“So there are two different lenses: one is as high up the abstraction ladder as you can get for as much of the services as possible. And then if you look at the software that you own, that would be in that PaaS layer, look at it across the spectrum.”
