Right now, most businesses are jumping on the cloud services bandwagon, moving workloads that typically reside on physical infrastructure into the public clouds, AWS and Microsoft Azure for example. These moves can be fantastic for your business and have enticing financial benefits, however, if it’s not executed well it can be a very painful and very expensive.
So how do you ensure your investment in the cloud delivers? Here are the 5 things you should consider before making the big move.
- Review your usage – this is the first step in identifying opportunities for cost reduction. In many cases were a migration was a ‘lift and shift’, a cloud instance may end up being over spec’d and significantly underutilised. By reviewing performance metrics, you should be able to identify opportunities to resize your instances to something more appropriate. Some cloud providers even provide services that will analyse your environment and provide recommendations for performance and cost optimisation (e.g. Trusted Advisor in AWS).
- Pre-purchase – Most public cloud providers allow you to pre-pay for a service for a fixed amount of time, such as 12 or 36 months. Pre-purchasing can bring significant discounts compared to pay-as-you-go, however, you need to fully assess your environment to ensure you get the most out of your purchase – you don’t want to be pre-purchasing resources that only have a short-term life. Depending on the provider, pre-purchasing of resources may also bring other benefits such as guaranteeing capacity within a given datacentre which is critical should there be a significant outage to a cloud service.
- Choose the right service – Cloud is a lot more than just a ‘virtual server’. Serverless architecture is now a reality and quickly gaining ground on traditional architecture due to services introduced in public clouds. Running web servers and databases is now possible without having any underlying server infrastructure to manage. This means costs can be calculated on a transactional basis rather than per hour for traditional virtual server hosting. Serverless architecture also greatly simplifies disaster recovery and backup, with many services including automatic backup options and multi-region replication.
- Scaling – Scaling up to meet demand and inversely, scaling down during quiet periods can bring significant cost savings. By having compute resources utilised when they are needed most you are not paying for ‘idle time’. Resource scaling signals how important appropriate monitoring is in a cloud environment. Defining appropriate metrics to monitor ensures that you have a complete picture of how your environment is operating and paves the way for an effective scaling policy.
- Automation – This is where the cloud platform really shines. By viewing the infrastructure as code rather than the physical hardware it is possible to configure and deploy complex architectures in one click. This ‘templatizing’ of an environment can reduce the effort required while also ensuring a consistent state. Automation can also be used for other cost savings, such as scheduling automatic power policies. For example, you could configure a policy to ensure your development environment is only powered on during business hours so you are only paying for resources when they are most likely to be utilised.
If you have just made the jump to the cloud, the steps above may seem daunting. So, if you need some guidance on your cloud journey contact the team at Starboard IT today.
If you’re looking for consultancy services for your next project contact Starboard IT to find out what we can do for you.
