AWS Migration Considerations Series: Why migrate to the Cloud?

 

Exploring AWS: Migration Paths and Operational Considerations

There are several ways that the hyperscale cloud environment of AWS works these days, from classical virtual machines, to containers, and serverless offerings. The assessment of which migration path works for you is covered in the article Migration Methodologies.

Much has been said about magical cost savings, performance, reliability and more, however , the truth is, all these advantages should be considered, architected, implemented, and then operated in the Cloud to be as magical as they sound. And that effort requires consideration; it does not occur by itself and is instead an operational cost.

But why choose AWS as your Cloud provider?

Historical long-term operational excellence

AWS started in 2006 with the Simple Queue Service. As we hit 2024, that will mark 18 years of continual operation of the services. The same APIs that were defined a decade and a half ago can still (generally) be used today.

The operational stability makes this dependable. And while the encryption protocols and ciphers used in transit may be updated to deal with today’s threats, older services can still rely on these systems.

Strong focus on security

AWS always says that security is job zero. Their security teams are proactive on maintaining their managed services, ensuring that critical exploitable vulnerabilities are addressed effectively. 

In 2014, when the Heartbleed vulnerability was breaking news, AWS managed load balancing service, Elastic Load Balancing, was patched worldwide for all customers within 24 hours. Meanwhile, there are still load balancers deployed outside of the Cloud that are susceptible to this. In 2017 a report showed there were still around 200,000 impacted devices. Using Shodan.io in 2020 still has a handful remaining!

However, security is a joint responsibility: the security of the Cloud, and security in the Cloud. AWS looks after the “Of the Cloud,” while the customer (or their managed services partner) looks after the “In the Cloud.”

You cannot take security for granted.
The shared security model is a key concept to understand. It is not a given that your virtual machine, that you have secured, is going to get patched and updated for you by AWS. There is an AWS managed service, but you will still have to patch and maintain your application on your virtual machine.

General applicability to a wide range of workloads

Having the AWS environment flexible enough to run almost any workload makes it applicable to nearly everything. Some cloud providers have niche or specific workloads they cater to, forcing you to stitch together distributed solutions from different providers. 

Commodity offerings, consistent world-wide

The Industrial Revolution has commoditized much of the IT infrastructure game. Just as universal access to electricity, facilitated by a grid, ensures that everyone has reasonably reliable power, the same principle has been applied to computing. It is now universally available, not just for computing but also for well-architected computing across multiple availability zones.

Data residency remains in the region you choose

It’s disappointing to reference this one, but some cloud providers would replicate your data worldwide or only offer durable storage if you choose another country to replicate it to. Both options are unpalatable to many organizations, particularly governments, who wish to have their content retained in their own jurisdiction.

Of course, should you select to have replication to other locations, this can be done (and even done as a managed service).

It is also worth noting some of the AWS AI services may copy data to other regions, as noted by Scott Piper, in order to “develop and improve AWS and affiliate machine-learning and artificial-intelligence technologies.”

On-demand flexibility, no commitment requirements

The ability to add and remove chargeable resources on demand, automated without having to negotiate, schedule, and wait is itself a miracle.

Large third-party software support

There is a lot of software that already supports AWS APIs, which makes solutions easy to deploy and maintain. 

History of price optimization and service innovation

Pricing is an interesting topic. You may not want to force your provider to underbill you, but the long-term profitability and longevity of the service you depend upon may be in question. But you do not want it to be gouging you, the client.

While we see efficiencies being made, it would be more helpful to share some of that with the client, then it makes sense to share some of that with the client, which is why AWS has applied reduced pricing to some services over time. An even more impressive factor is that More impressive is that the price drops are automatic, with no client action required to benefit. This breeds client loyalty and often feeds more demand.

Continuing innovation

Cloud providers that don’t continuously improve their offerings, modernizing and securing the parts of the technology stack that are their responsibility, are frozen in time. For example, a CDN service that today does not offer the current TLS 1.3 protocols for ensuring the secure transport of data over untrusted networks will soon be eclipsed by more nimble competitors.

Likewise, a CDN service not restricted to only the latest protocols will fail compliance requirements for many organizations. 

Strong providers in the Hyperscale Cloud space introduce new capabilities without any immediate negative impact on currently deployed customers. Managed seamless transitions in technologies—sunrises, transitions, and sunsets—are key to stability and customer satisfaction. 

Industry analyst reports

For the last 15+ years, Gartner has ranked AWS as the leading hyperscale Cloud provider. Forester Wave articles consistently call out various AWS services amongst the strongest in their field. 

The level of detail that the analysts go through with the providers is extensive, covering all areas of operations, service development, security, and much more.

While these reports alone are insufficient, they help to confirm the capability to date, giving confidence to the consumer.

You have a lot to think about when choosing a Cloud provider. And While many will show you the golden land of milk and honey with instantaneous results, the reality is that  successful migration to the cloud takes time, effort, and attention to detail that can only be done with focus (and experience). Tune in next time when we discuss the migrations that happen once you have moved to the cloud.

Akkodis has been an AWS Consulting Partner since 2013. Learn more about our AWS Practice and services.

By James Bromberger, VP Cloud Computing, Akkodis Australia