.png)
A prominent company was facing a significant challenge with their flagship product. Their Azure cloud infrastructure, consisting of 60 Class D Virtual Machines (VMs), was incurring a hefty monthly expense of $50,000. The primary issue lay in the memory usage, which was constantly hitting the upper limits, averaging between 80% to 100%. This heavy memory utilization triggered frequent auto-scaling events, leading to the deployment of additional VMs and further escalating costs. Despite this, the compute utilization remained surprisingly low, averaging only 20% to 30%, indicating a mismatch between the selected resources and actual computing needs.
Clear FinOps stepped in to conduct a comprehensive analysis of the current setup. Our investigation revealed the core issue: the company's reliance on Class D VMs, which, while sufficient in computing power, fell short in memory capacity for their specific requirements.
To address this, Clear FinOps proposed a strategic shift to Class E VMs. These memory-optimized machines offered over double the memory capacity compared to the Class D VMs, without a significant compromise in computing power. This change was crucial as it aligned more closely with the company's actual usage patterns, where memory, not compute power, was the bottleneck.
The transition plan also involved reducing the total number of VMs from 60 to 35. This reduction was possible because the Class E VMs' enhanced memory capacity adequately handled the workload that previously required a larger number of less efficient VMs.
The shift to Class E VMs yielded remarkable results:
In conclusion, the strategic intervention by Clear FinOps not only led to significant cost savings but also optimized the company's cloud resource utilization, demonstrating the value of aligning cloud infrastructure with actual workload requirements.
Explore Clear FinOps Services for more expert guidance and support in mastering Azure cost management.
.png){kind=spacer}