While it’s easy to model out pricing for your public cloud usage, you cannot plan for high, unexpected spikes in demand.
Despite the widely accepted view that public cloud computing saves money, these spikes add hidden costs that can dramatically reduce the return on your cloud investment.
“Cloud computing doesn’t always save money,” wrote Tom Nolle, president of CIMI Corp., in a recent article.
The public cloud cost debate is a challenge for CFOs, CIOs, CTOs and even venture capitalists across the technology industry. That’s why it’s so important to take back control over cloud usage for your baseline workloads, particularly your high-data, high-bandwidth applications. Let’s look at some examples where spikes in demand threw public cloud pricing models out of whack.
Zoom
When the COVID-19 pandemic hit and forced so many people to work from home, the videoconferencing platform Zoom experienced a massive increase in usage — and had to significantly increase its cloud spend to keep up.
In addition to beefing up its data centers, Zoom’s engineering operations team added servers and other equipment inside every one of the company’s data center locations and increased capacity with its cloud infrastructure providers.
Any service, such as videoconferencing, that relies on large data transfers may be particularly susceptible to high unexpected cloud costs. Cloudflare recently reported that Amazon Web Services (AWS) charges customers in the United States and Canada an 80X markup — not 80%, 80X — on data egress.
Pinterest
It’s not just unexpected pandemic pricing that catches enterprises off guard. Underestimating peaks in cloud usage throughout the year is costly as well. Pinterest, which had paid AWS in advance for services, had to buy additional capacity at a higher price during the 2018 holiday season. The company spent roughly $190 million on AWS that year — $20 million more than it had initially expected, according to a person close to the company.
These unexpected costs can hit established enterprises especially hard, as Andreessen Horowitz partners Sarah Wang and Martin Casado recently pointed out:
“As industry experience with the cloud matures — and we see a more complete picture of cloud lifecycle on a company’s economics — it’s becoming evident that while cloud clearly delivers on its promise early on in a company’s journey, the pressure it puts on margins can start to outweigh the benefits, as a company scales and growth slows.”
Taking Control Back
If you want to take back control over your costs, you need the amenities of public cloud in a private cloud environment for your baseline workloads. A key property of public cloud is that it is supposed to be elastic - scaling up and down with demand - and therefore, isn’t always well suited for baseline workloads.
Platina Command Center brings Infrastructure as Code (IaC) simplicity and flexibility to the private cloud, enabling on-premises orchestration of network, servers and storage, just like in the public cloud. In addition, Platina provides on-demand access to bare metal compute, containers with Kubernetes orchestration, software-defined storage with Ceph, and automated networking. With Platina, CTOs and DevOps teams can enjoy public cloud conveniences, while CFOs and CIOs enjoy predictable infrastructure costs.