Amazon Web Services had an outage yesterday. With such widespread cloud migration planning and adoption taking place these days, public cloud outages receive a lot of press, so you likely already knew this. People will want to know why it happened, but while that is important to Amazon, it is not important to the rest of us.
What is important is that you think about these scenarios when designing your workload for deployment in public cloud. You cannot prevent outages like this from happening, and what happened yesterday isn’t an AWS-specific problem – yesterday it was AWS down, the week before it was someone else, tomorrow someone else again. But by understanding your appetite for down time and all the capabilities made available to you from the public cloud, you can design your workloads so that outages like this have less impact for you.
AWS provides great guidance on reliability as one of the five pillars in its Well Architected Framework. Many of these principles and best practices have been around IT for many years, but with the right skill sets, knowledge, and partners, are made easier to implement and maintain using public cloud. Services like automatic recovery, managing change in automation, and horizontal scaling are good examples.
I should also point out that there are four other pillars that are equally important – operational excellence, security, performance and cost optimization, but those are for another day.
The Design Phase is a Critical Component of the Migration Project
I like to stress with customers the importance of the design phase when considering a workload migration to cloud – every minute spent here will pay for itself multiple times during its run lifetime. Working with the technical and business stakeholders is critical and will result in a much more robust design and success of your migration.
With successful cloud implementation, addressing all of this begins during the design phase of your cloud migration and architecting based on your business needs. Do you need to scale quickly? How much downtime can you afford? Do you need geographical redundancy? These and other questions should all be answered, incorporated into your design and understood by your stakeholders.
Just like everything else in IT, ultimately migrating to and operating on cloud well comes down to planning and executing from a clear starting point and roadmap, as well as leveraging the best of your internal resources and partners for every phase of your IT workload’s journey. The best way to avoid “two steps forward, two steps back” in the cloud is to start with a real understanding of your IT environment’s structure, dependencies, and requirements by executing a cloud assessment. There are a variety of methods and tools available for this kind of exercise. When you combine this critical knowledge with partners and resources that offer real cloud experience and know-how, you can not only avoid these pitfalls, but leapfrog your competitiveness and customer reach.
Pivotree Can Help with Your IT Strategy and Cloud Migration Planning
Pivotree’s experience with running critical infrastructure in any environment means that we can offer the insight required to build and execute a holistic IT strategy that provides both reliability as well as the business value you need to reach customers while controlling costs and staying ahead of the competition. Talk to us and let us learn about your environment and challenges; we love to apply our knowledge and experience to deliver real cloud outcomes to our customers.