Is your cloud infrastructure ready for ‘peak bloat’ season?
Published by
Daemon
on
It’s an exciting time of year. Black Friday, Thanksgiving, Amazon Prime Day and Christmas (and the January sales) are a high-traffic golden period of sales for retailers. With the surge in demand comes a massive test of business systems.
We spoke to Daemon’s own Cloud Principals, Marc Peiser and Harry Azariah, about what ‘bloat’ means at this time of year and how to avoid unnecessary stress on your business.
What is ‘peak bloat’ in cloud terms?
‘Peak bloat’ is the season when businesses invest, sometimes excessively, in cloud provision. Reasons for this can include:
- Higher numbers of staff working ‘remotely’ during the festive break
- Scaling to meet increased customer demand
- Adding resilience to ensure reliability of key systems in a period where the teams that usually support them are depleted.
In 2023, Amazon spent roughly the equivalent of $102 million on cloud provision in preparation for Amazon Prime Day. Although this number is estimated based on the listed retail pricing for that infrastructure, it’s still staggering. This might sound excessive, but the return on investment was huge - Amazon took over £12.7 billion in sales in 48 hours.
For Daemon, it’s time to be alert and responsive, when major changes to the infrastructure are paused and we closely monitor cloud infrastructure.
“As cloud engineers, our priority is to stay alert to what’s happening in real time, ensuring we can quickly identify and notify businesses of any issues before they escalate.”
Marc Peiser
Retail is particularly stressed during the Holiday period. It’s a great time to make record sales, but it’s important to not over-invest. Businesses with proper, cost-effective architecture can easily avoid ‘bloating.’
“Lots of companies invest in cloud provision at this time of year. Infrastructure and application teams will be running lean as people take time off to enjoy the festive period. Businesses must be still able to provide the same level of service that employees and customers enjoy throughout the year.”
Harry Azariah.
Connect with Harry on LinkedIn
What are the risks of not preparing for high seasonal demand?
Slow, lagging page load speed is a huge problem between November and January. It’s a big driver in lost sales and in addition, reputational damage. Customers who aren’t served in a timely manner are more likely to visit a competitor. If systems do fail, that’s also a big problem in terms of lost sales. During peak periods when the demand is even higher, or special offers cause a logjam, this is exacerbated. For every minute your service is down, you could be making significant losses at this time of year.
Performance and responsiveness are vital for eCommerce. Over a quarter of total retail sales happen online, and 70% of smartphone users use their devices to buy. Therefore, the risks of not preparing for increased seasonal demand are high.
When thinking of high seasonal demand, we should also consider the fact that many will be working from home rather than traveling into their offices, without this provision businesses may have to close over the period. The beauty of the cloud makes it easy to provision additional infrastructure to support this without having to procure new infrastructure. The downside to this is, left unchecked, costs can quickly spiral.
Scaling to meet demand is critical. Luckily, one of the best things about having a cloud architecture is its elasticity. But of course, you don’t want to over-commit or overspend. How can you optimise your provision?
How can you reduce Peak Bloat?
In past years, businesses might have invested in physical infrastructure, but in 2024 the investment is spent on cloud elasticity. Scaling your cloud provision in preparation for demand is good business sense, as long as your apps are architected correctly, you should be able to scale dynamically with the right cloud support.
- Conduct a Well Architected Review
One of the best ways to prepare for seasonal peaks is to scale strategically. An AWS Well Architected Review will give you tailored insights into weak spots:
“Well Architected Reviews cover reliability, performance and operational excellence. The actionable insights they deliver act as a tailored roadmap, helping you scale efficiently and effectively while avoiding unnecessary costs.”
Marc Peiser
- Auto-Scale
Another smart tactic for preparing but not over-preparing is to implement automatic scale provision. This means your cloud provider offers more only when you need it, preventing unnecessary spending.
How Daemon can help
Daemon is a proud Microsoft and AWS Partner. Our Cloud team are experts in the Well Architected Review - a quick, thorough way of testing your cloud provision and finding key areas for improvement. We also offer a Cost Optimisation service for Azure and AWS, enabling you to see where savings can be made.
A Well Architected Review highlights areas of focus across several key pillars including reliability and operational excellence. The findings will show where you need to improve, and are rated by criticality, enabling you to focus on areas of high risk.
“If you’ve experienced outages or performance issues, you can seek help. We’re here to ensure you provision for scale and avoid problems in the first place. Post-peak is a great time for a performance review, so get in touch with us if you’ve had any issues.”
Harry Azariah
January 2025 is a useful time for many businesses to reflect. Did you overspend during the festive period? It’s a good time to reach out and explore ways to improve your cloud architecture, and Daemon are happy to help.
For this and any other cloud performance issues, get in touch with Daemon today