Multi-cloud strategy: A checklist of things to keep in mind
When implementing a multi-cloud strategy, it is well worth going into it with both eyes open by planning ahead to face potential challenges and setting a few long-term goals.
In part 1 of our series on multi-cloud, we discussed the four common stages in a cloud adoption journey. In part 2, we answered some common questions from our customers regarding multi-cloud. In this concluding post in the multi-cloud series, we wish to warn of potential pitfalls.
The advantages of having a multi-cloud strategy are undeniable. Yet, adopting a multi-cloud strategy requires certain due diligence. Here are things to keep in mind.
Make sure multi-cloud doesn’t mean multi-governance
Enforcing policies equally across multiple cloud systems can be complex. Enterprises must develop specific systems and practices to ensure streamlined deployments while maintaining consistency and performance standard across systems.
One way to do this is by enabling automation wherever possible. For example, streamlining deployment with tools such as Spinnaker, Chef, Puppet etc. A success story in this regard is Netflix, which saw a doubling in its release cadence after implementing Spinnaker, their own custom cloud deployment orchestration engine, which is now open-sourced. Andrew Glover, engineering manager at Netflix, sums Spinnaker up by calling it “a paved road where best practices are codified in a single platform.”
Upskill your teams
A top-notch IT department will have to be maintained to oversee the multiple infrastructures and how they function together as a cohesive whole. Avoid services which require major reconfigurations, such as a cloud-specific adaptation of an application, as deployment or migration will be inefficient and take a huge amount of work each time.
Most large IT teams eventually develop a multi-cloud strategy, especially when responsible for both corporate IT solutions and customer products. You can build up to it and ensure a smooth transition by maintaining commonality across departments where possible — like operating systems, sysadmin tools, and development languages and frameworks, for example.
So, a big plus to a smooth-running multi-cloud strategy? Standardisation of tools and processes, especially around scalability and deployment, for a flexible and effective DevOps team. Also, compare the benefits of building a team internally and finding a reliable managed services partner.
Secure your information, twice over
Companies working across multiple cloud systems must develop strict policies that can be implemented and followed by users uniformly across their organisation, no matter the platform or vendor. This can be challenging, depending on each system’s capabilities. There must also be a solid contingency plan in case of a security breach or data loss. Failover testing needs to happen multiple times a year, as security is an ongoing concern. Enable automatic switching to a backup cloud in case of primary cloud failure and take measures against data loss.
While cloud vendors have multiple security tools and protocols in place, your data’s security ultimately remains your company’s responsibility. As such you can’t afford to make assumptions — there has to be a system in place that tests that everything performs as it’s supposed to. Multi-factor authentication and Zero Trust security models should be a matter of course, not just for workloads involving sensitive information. Especially in cases where a single engineer or team wholly manages an infrastructure, least privilege access, and multi-factor authentication go a long way towards ensuring your system’s stability and security.
Never miss a bill
Keeping track of different cloud providers for different platforms and departments can be very challenging. Especially when costs keep fluctuating due to scaling, changes in vendors, and so on. Given its pay-as-you-go nature and ease of starting up new resources, cloud consumption can easily go over budgets if left unnoticed — out of sight will be out of control!
For maximum cost-efficiency, clear visibility across virtual machines and corresponding projects/developers/departments etc. is essential. ROI on each cloud platform should also be tracked on an ongoing basis.
Our 24×7 managed services teams monitor and analyse cloud consumption optimizing costs regularly.
Get on the right side of the law
When you have a sprawling, multi-cloud, global organisation, it can be difficult to be aware of where different data physically resides. Not knowing where your data resides puts you at risk of violating multiple regulations, including the General Data Protection Regulation (GDPR), which Forbes recently called “the greatest change to European data security in 20 years.” For example, GDPR states that you must inform customers of a data breach within 72 hours of learning of an attack — hard to do when you don’t know where your data is.
Forming a solid data governance strategy should be a priority to every cloud-using organisation that wants to avoid potentially serious consequences to a misstep. It will also make it easier to respond to new regulations or compliance requirements as they come up. The best way to do this is to institute a Chief Data Officer and a dedicated department for data governance. This will help in maintaining data integrity, implementing processes and systems for data flow. This will also ensure that the organisation remains compliant with various data governance laws.
While the multi-cloud strategy is the most flexible, adaptable and cost-effective way to go, it involves some strategic planning and overseeing to avoid challenges and potential pitfalls. Nothing you can’t overcome with a good cloud partner.
1CloudHub has helped multi-national enterprises move to the cloud without a glitch. We can help you. Speak to a cloud consultant today.
Updated on June 5, 2020
Sharing is caring!