Multi-Cloud approaches can protect you from downtime, cyber-threats, and improve your bottom line

26 May 2023, by Slade Baylis

With reference to IT, when people hear the term "cloud” they often interpret it in many different ways.  In fact, when the US government asked the National Institute of Standards and Technology (NIST) to come up with a definition, its definition began with a caution stating that “cloud computing can and does mean different things to different people.”  However, for those not already familiar with it, most people will be visually drawn to images of its namesake - that of the ethereal clouds that adorn our skies. 

Given this, many will assume that when it comes to the "cloud" in the IT world, any services built on it will be just as enduring and won’t be as prone to hardware issues that affect more typical IT approaches such as running services on one's own physical servers.  However, back on the ground where the rubber meets the road, cloud services still experience outages that can take the services hosted on them offline.  The reason for this is simple - they aren’t ephemeral.  In fact, they are very much mechanical, running on massive servers, powered by generators, and cooled by air-conditioners and water-cooling.  And this is why it’s important to factor these sorts of failures into your risk-assessment when designing the infrastructure of your systems.

In a term that’s growing in popularity, more businesses are moving toward “Multi-Cloud” approaches in order to mitigate these risks.  At a high level, this strategy encompasses using multiple different vendors in order to reduce risks of downtime, limit the potential damage of cyber-threats, and position yourself in a way that allows you to leverage flexibility.  Furthermore, most large cloud providers (AWS, Azure & Google) are expensive at scale, so using different Multi-Cloud strategies such as complementing colocation services can also improve your bottom line.

Multi-Cloud and Hybrid Cloud – How are they different?

A common term used when speaking about cloud infrastructure is that of Hybrid Cloud – which is the practice of utilising both your own private systems and public cloud technologies in tandem.  The prevalence of this term is understandable, as most businesses go through a phase of using a Hybrid Cloud approach as they move from their more traditional IT infrastructure (such as using physical services that are hosted on-premises) to using cloud-based replacements. 

However, a term that people are generally less familiar with and often mistakenly use interchangeably with the above term is Multi-Cloud – the description of which is very different to Hybrid Cloud.  In basic terms, Multi-Cloud simply means “multiple clouds” -  so any company that hosts their applications on different cloud providers simultaneously is using a Multi-Cloud approach.  Even companies that use multiple different SaaS tools (such as Slack, Salesforce, and Dropbox) are included in this definition. 

Multi-Cloud deployments utilise the services from multiple different cloud providers in order to host their workloads – such as applications, websites, and other services.  They do this in order to increase resiliency; avoid being locked into a single provider; to allow for quicker disaster recovery; as well as to attain better pricing through the leverage gained from the flexibility of their systems.

Avoiding vendor lock-in and leveraging flexible infrastructure to get better pricing

When progressively moving from more traditional IT infrastructure to having your systems hosted in the cloud, this by definition requires you to rely more and more on those external cloud providers.  When using a single provider, this can result in it becoming more difficult to move away from them, should you ever want to in the future – and this is what’s referred to as “vender lock-in”.  Whether it’s due to pricing increases, unsatisfactory support, or anything else, most businesses would prefer to make sure they have the flexibility to move if they want or need to in the future.  Due to not having all your eggs in one basket, Multi-Cloud approaches provide the flexibility needed to achieve this – as you are able to rapidly redeploy your systems to an alternative cloud-provider’s infrastructure - if the need arises. 

One way to achieve this specifically is through the use of cloud-to-cloud backup technologies, as we’ve covered this in our recent Cloud-to-Cloud Backups Why backups of your data are just as important as ever! article.  In short, most cloud-services don’t actually come with any form of backup at all!  However, even if they do, backing up your systems with an alternate cloud vender can allow you to restore your systems quickly in the event of an emergency - this is why we recommend they be included in any organisation's DR (Disaster Recovery) planning.  As an additional benefit, doing so can also allow you to easily move your systems permanently if you become dissatisfied with your primary provider.

Another way to allow for that rapid redeployment is through the use of containers or microservices.  These technologies allow applications to be portable, often permitting them to be deployed or redeployed quickly onto different FaaS (Function-as-a-Service) or CaaS (Containers-as-a-Service) clouds as needed.  This helps businesses avoid being locked-in with a single provider, giving them the flexibility to move in response to availability issues, ever-changing legal requirements, or even to take advantage of better pricing!

On this, one of the hidden “gotchas” with most of the large cloud providers - such as AWS, Azure, and Google – is that it can be cheaper to get going initially, but as you scale and require more compute and resources, the cost can increase dramatically, even with reserved pricing.  In basically all cases, it eventually becomes more cost effective to colocate your own server and contribute it to a cluster (like we offer at Micron21).  By contributing your own hardware to existing cloud infrastructure, you are able to maintain the redundancy and other benefits of cloud-hosted services whilst benefitting from the cost-savings gained from utilising your own hardware.

Increased flexibility also has another advantage, which is the leverage it gives you over your provider in negotiating on price.  If there is a large amount of “inertia” with your systems – i.e. moving your systems to another provider would require monolithic effort and be a costly exercise – it’s more difficult to negotiate with your current provider for a better deal.  However, if you are able to move your systems at a moment’s notice, you’re on much firmer footing to get the best price possible for your infrastructure!

Increased resiliency and data protection, though with additional security challenges

It shouldn’t be much of a surprise, but avoiding having all your eggs in one basket – i.e. by having your workloads hosted across different cloud providers - allows you to be able to avoid the risk of all of your systems having an outage due to provider-side issues that you have no control over.  In fact, with sophisticated enough configurations, you can even load balance your systems in such a way to avoid downtime entirely in these instances.

Data sovereignty is also another factor to consider – through being able to host systems with different vendors, you are able to choose which locations to host specific data to meet any regulatory requirements.  For example, for Australian government departments, making sure data is stored within Australian shores is often a critical and an unnegotiable requirement.

With regards to cyber-security threats, such as ransomware, the segmentation gained through diversifying your IT infrastructure can also potentially limit the impact of any incident in the unfortunate event that they occur.  Whether it’s through hosting different systems with different providers or through backing them up to an alternate vendor’s platform, you’re able to protect yourself from losing your own or your users’ data, as well as prevent you from being compelled to fund further cyber-crime through payment of their demanded ransom.

However, also mportant to keep in mind, is that there may be some potential drawbacks to using a Multi-Cloud approach.  Primary amongst these is the fact that using multiple different providers to host one's services means that an organisation will have a greater “attack surface”.  In simple terms, having a greater attack surface simply means that the more software, hardware, and systems one uses, the larger the likelihood of them containing a vulnerability that could be exploited.  Similarly, simply the complexity involved in managing, monitoring, and controlling access to many discrete systems can introduce additional cyber-security risks that wouldn’t be present with single-vendor configurations. 

Whilst there are advantages to using multiple providers - due to these potential risks, it’s important to be judicious about the cloud services you use – with the aim being that you should use multiple providers where it’s beneficial and consolidate your different tools where it is not. 

Gaining more control and freedom over your infrastructure 

Overall, the common thread here is that through utilising the services of multiple cloud vendors, you are able to give yourself the freedom to move your applications as needed.  Whether this in response to price, latency, geographic location, downtime, regulatory requirements, or just personal preference - reducing your IT inertia gives you more options and flexibility to meet your objectives, whatever they might be.

There are always benefits and drawbacks to any approach, but Multi-Cloud approaches should always be included in any discussion about security or reliability, or even just conversations about your organisation’s IT budget.

Interested in finding out how you can best leverage a Multi-Cloud approach?

If you would like to find out what a Multi-Cloud approach could mean for you, reach out to us! 

You can call us on 1300 769 972 (Option #1) or email us at sales@micron21.com to find out more.