Recently, our company has become the first AWS Well Architected Partner in Central Europe – we’ve even started to offer AWS Well Architected Reviews – but we noticed that the nature of Cloud Architecture is one that is seldom discussed openly.
Thanks to our experience and established Cloud Architecture services, we wanted to share some knowledge with you. We want to answer some of the key questions – such as why your Cloud architecture matters and what can be done to improve it – right here.
So, let’s get started!
A Foundation For The Future
They key issue with Cloud Architecture is that it’s often the foundation for how anything else built on top of it performs. It’s this Architecture that connects microservices or ensures an efficient resolution for your end users. Problems or inefficiencies here can easily spread.
To this end, it’s important for businesses to understand the need for a strong, updated architecture and – just to be clear – an established understanding of Cloud Architecture as a whole.
What Is Cloud Architecture?
When you have a number of components in the Cloud, you likely have a need to connect or support them with additional subcomponents. Once you do this, you start having the beginnings of Cloud Architecture.
In short, Cloud Architecture doesn’t just refer to the various components you have on the Cloud. It also includes the connections and relationships between each component and the various subroutines that are involved.
For example, if you are analysing data, you likely have a number of databases stored in the Cloud (and you should – the Cloud is fast becoming the industry standard for Big Data operations), but this is supported by various processes to edit, transform and interpret this data in various meaningful ways. Depending on your use cases – such as internal analytics or limited end user access – you may have even more components involved. All of these need to interact in different ways.
How you build this and support these interactions is a key part of Cloud Architecture. It sits underneath all your other solutions – so it’s important to get it right.
Why Does It Matter?
For companies in the Cloud, their architecture is the foundation for their products or solutions. Issues here cannot be overlooked. Using the previous example, if there are issues in how data is sent to other components or modules, this can disrupt the entire service.
Perhaps its performing too slowly, or perhaps there’s a security issue during this process? No matter how optimised the individual components may be, it only takes one weak link to introduce disruptions.
On the other hand, it might also be possible that your architecture is safe and secure. However, if it hasn’t been updated in some time, is it as optimal as possible?
Cloud providers like AWS are introducing new services all the time, as well as improving old ones. As a result, companies that take a look back and their existing structure often find their solutions are outdated and a cheaper, more efficient solution is right within their reach.
AWS Well Architected Framework Pillars
As companies move forward with the Cloud, they are often building new solutions or services, but failing to look back and update the existing architecture. As we’ve already mentioned, this can be an issue. Perhaps it’s not optimal, or perhaps there are lurking issues that simply haven’t caused a large enough issue (yet!) to get noticed. It’s also possible that you found an issue, but misdiagnosed the root cause.
This is why Amazon Web Services (AWS) launched their Well Architected Framework. It’s designed for the specific purpose of analysing existing infrastructure to identify both key areas of improvement and critical risks.
Fortunately, we’ve got plenty of experience in creating modern, innovative Cloud Architecture – our work has even appeared on AWS’ own “This Is My Architecture”!
But what exactly does the AWS Well Architected Framework look at? It specifically looks at five key areas:
Cloud security is always an important issue. With businesses looking to migrate to the Cloud on an increasingly larger scale, there’s always a need to stay secure. While Cloud providers such as AWS go to great lengths to secure their own platforms, it’s always possible to leave unintentional exploits or issues within your own setup.
There are numerous possible exploits or situations to consider and it’s likely that not every angle has been considered. This includes:
Widely open permissions – could a lack of restricted access allow undesirable users into areas that they should not be able to enter?
Exposed secrets – Are secure areas of your architecture truly secure? If a user can gain access to information they shouldn’t be able to access, this can be a serious breach of data privacy.
Privilege escalation – While a user might have access at a certain restriction (such as read-only) they may still have access to either the information or the direct means to increase their privileges. Such an issue is a big cause for concern, as someone with limited access may very well end up with administrative roles.
Of course, these are just a few examples. Because we know important security is nowadays, our experts go above and beyond standard procedure to make your Cloud infrastructure as secure as possible. In short, we will look at your architecture from every potential angle, with no stone left unturned.
Ultimately, everything you do on the Cloud serves to achieve your business goals. Yet this is one area where few companies are truly making the most of the Cloud, automated services and the ability to respond near instantly.
Previously, companies relied entirely on manual processes, which meant that updates were large, slow and infrequent, with little time to learn or document the best strategy. When moving to the Cloud, organisations often seize many of the benefits, but this old fashioned approach still remains. Operational Excellence is the opposite of this, embracing the Cloud for faster, more agile results.
During this part of the review, we check to see if your architecture is able to support frequent, small and reversible changes to your workload – this is essential for staying adaptable. The review process also checks if you perform operations as code, if you utilise annotated documentation to keep up with changes and if you anticipate both success – as well as failure – in your infrastructure, with regular testing to ensure it is reliable.
We’ll also look to see how you refine your own operations and procedures as well. Are you hosting game days? These are an important way to test your procedures are effective, that your teams are familiar with them and your business can tackle such challenges with ease. With any given scenario, teams should know who to contact and how to proceed as efficiently as possible.
All of these features ensure businesses can prepare, operate and evolve. They adapt quickly, but learn from both success and failure to constantly and frequently refine their solutions.
Downtime is the enemy of virtually any business and, unfortunately, there are many ways in which a system can become unreliable. Data loss or errors in one component can then cause others to become unusable, all of which drives customers away as business values start to disappear.
On the plus side, there are ways to anticipate, plan and prepare for such disasters. This pillar looks for these solutions, such as recovery planning, cross project operations and backup processes, if necessary.
An underperforming solution will struggle to derive key value. After all, if your product is slow, struggles with larger workloads or simply doesn’t operate as intended, this poor performance will become an urgent issue for your company overtime.
Often, these issues are caused by choosing the incorrect components or configuring them the wrong way. Resource allocation often needs to be fine tuned to ensure that the right amount of resources are provided when needed, whilst still not wasting resources during periods of reduced activity.
Scalability is also a vital issue here – are your solutions performing perfectly at smaller scale, but struggling to maintain such service as data volumes, users and tasks increase?
Finally, cost optimisation is often one of the key reasons people moved to the Cloud – yet it’s often not considered again. As your infrastructure is built and expanded, it’s very easy to create improper solutions that aren’t to today’s standards. As a result, while you are still saving money vs on-premise solutions, you’re likely still losing out in comparison to what you could be saving.
Wasted resources, as mentioned in Performance Efficiency, is a classic example. If every task uses the entire server, rather than smaller computational processes, costs can skyrocket. This can be clearly seen in industries such an eCommerce, where usage can vary wildly. A scalable, serverless approach ensures just the right amount of resources at each moment.
In fact, we’ve done a number of PoCs and experiments with scalability, including containerised applications, and we know that some solutions are cheaper than others. Finding the right balance between performance, scalability and cost is essential to truly getting the most benefits from the Cloud!
Cloud Architecture Best Practices
As mentioned earlier, the Cloud is a constantly shifting and evolving landscape. As such, best practices often change and this is one of the reasons why Cloud architecture reviews are so vital. Without these retrospectives, such issues continue to grow and grow as they become more outdated.
For example, scalability will always be a strong benefit of the Cloud, but it’s often not the first thing companies considered when they moved over many years ago. Now, with the ability to fine-tune resources to spin up on demand, companies can cut back on resource expenses and ensure their solutions stay as optimised as possible.
On the other hand, if they don’t do this well, any subsequent service that connects with these older components will cause these resource-heavy sections to either eat up their limited processes or spin up entirely new, un-optimised cycles. Either case – poor performance or increased costs – is unintended, unwelcome and, in today’s business world, unacceptable.
Aligning With Your Business Goals
At the end of the day, the Cloud is like any other business tool or solution. If it’s not directly helping your business goals, then it’s not helping at all.
This means that businesses should pay attention to exactly what they want to get out of the Cloud. For example, if you’re working with large sets of data and want to get the most insights, the Cloud Architecture for Big Data will look very different from, let’s say, an eCommerce solution. The former wants to store data and scale up analytical tools, while the latter wants to handle numerous instances at once. Both focus on scalability, but in different ways.
In fact, this is where many of the AWS Well Architected Framework Pillars can prove the most useful. By improving efficiency, you can drive better value straight to your end users, for instance. Each factor has its own direct benefits but, combined together, the advantages compound all the more.
If a solution is failing in even one area, this can be the minor difference that causes customers to choose a better solution – especially in a competitive market where many options are readily available.
In short, having the right architecture is essential for any business. A poorly optimised framework might not spell doom for your company, but it will mean you’re missing out on some of the newer, more modern benefits of the Cloud. That said, there are also some issues – such as security – that can simply never be overlooked.
It’s better to take the time now and take action, rather than waiting for the worst to happen. In short, the sooner you assess your current architecture, the sooner your business can refine, improve and enjoy the benefits. In fact, such reviews can be accomplished quite quickly, giving you immediate insights that enable you to take action in mere days!
So why not consider a Well Architected Review and get started?