When an organisation migrates to the public cloud, optimising workloads is a vital review business must undertake to fully realise the Cloud value proposition and reap gains from investing in digital transformation.
However, the time and skills to perform a Cloud utilisation analysis and re-architecting for costs may not be an in-house capability in many businesses.
- Further reading How to achieve the best cost optimisation on the public cloud.
Engaging a trusted AWS consultant to perform a Cloud investment strategic cost assessment will help an organisation gain the agility benefits from moving off-premise while adhering to a budget.
By performing a Cost-Optimisation Assessment (COA), an organisation receives the recommended actions it needs to take and provides the estimated monthly savings accrued by focusing on areas of high consumption and mitigating them based on four Cost Optimisation Pillars of the AWS Well-Architected Framework.
These issues are covered off in an ebook from AC3 called How to achieve the best cost optimisation on the public cloud.
These focus areas are expenditure awareness, cost-effective resources, matching supply with demand, and optimising over time.
The objectives are achieved using Cloudhealth to improve Cloud visibility, simplify cost management, increase security, and govern multi-Cloud environments by automating governance policies and uncovering security risks before they become problems.
The Cost-Optimisation Assessment is the starting point for helping customers realise their full transformation potential by migrating to the Cloud. One customer this helped had been commissioned to transform their compute environment from a legacy in-house on-premises data center to a fully managed private Cloud-based offering.
This customer had an ever-changing need to innovate and provide more effective solutions to meet its client’s needs. To do this, the firm leveraged multiple bespoke and off-the-shelf solutions to offer a suite of products and services running on a vast array of hybrid compute, delivering various customer requirements.
This demand required them to be agile, flexible, and elastic. The ability to bring products to the market faster and scale capacity based on customer utilisation patterns is central to innovate and deliver continually.
This meant the biggest challenge facing this customer was an inability to forecast their AWS usage and monitor costs effectively. To overcome this problem, the company’s AWS spend was analysed to uncover any potential cost savings.
A detailed assessment of their workloads was performed based on a three-stage approach. This included maximising cost savings and realising quick wins within the business, moving its large traditional datasets to AWS — using services such as Amazon S3, Storage Gateway along with connectivity through Direct Connect and Transit Gateway.
The COA identified a potential 23 per cent savings for this customer. This demonstrates how cost transparency can help organisations to manage their Cloud costs and allow executives to have better conversations with internal teams around business value — making the teams accountable for their cloud spending.
The second stage was to use the most cost-effective resources available that align with the customer’s needs. The assessment revealed the customer’s top cost drivers were EC2, EBS-Storage, Workspaces and RDS. These are the most common cost drivers for many AWS customers.
By using rightsizing tools, it was discovered that approximately 60 per cent of this customer’s EC2 instances were from previous generations. From this finding, the recommendation was made to migrate to the newer generation of AWS instances, which would save the customer seven per cent annually.
Furthermore, the assessment showed this customer could also save an extra 23 per cent if they converted these instances into Reserved Instances (RI).
The rationale behind these rightsizing recommendations is that upgrading instances to a newer generation is consequently cheaper. It also ensured that the recommended instances could handle the same amount of usage.
The final stage in the process involved matching supply with demand. Therefore, scaling and scheduling formed a critical recommendation put to the customer.
This is because customers needed to stop paying for what they use and start paying for what they need. This can be done by implementing an automated solution to turn instances on and off as needed.
Another recommendation for this customer was to consider using auto-scaling groups to adjust the resources dynamically according to the needs of the workload. This meant that the customer would only pay for the instances that were being used.
This resulted in driving down costs by launching instances when they were needed and terminating them when they were not.
Additional steps can also be identified to reduce wasteful spending, including deleting unattached EBS volumes, deleting snapshots, deleting disassociated elastic IP addresses, and Instance modernisation.
Thus, after performing a Cost Optimisation Assessment, this customer was able to operate in a highly flexible, scalable, and secure AWS environment that allowed them to build and release new features faster and reduce the management overhead cost that their previous on-premises solution required.
This article is published by Which-50’s Digital Intelligence Unit (DIU) on behalf of AC3. DIU Members pay to share their expertise and insights with Which-50’s audience of senior executives.