How to use metrics to bring your cloud costs under control
Cloud cost control is an essential part of corporate cost management. IT administrators need to report on budget usage on a regular basis, however, they often face problems such as lagging perceptions of capital risks, complex and time-consuming data analysis pipeline building, and customized data analysis requests from various departments.
To solve the above problems, a company chooses Kyligence Zen as the metrics store for cloud cost management. After a period of operation, cloud costs dropped significantly and eventually stayed at an expected level. In this blog, we will use this real case to demonstrate how to build a metrics store for cloud cost management in just two steps, to offer insights for the daily monitoring and the retrospective analysis.
Create unified metrics for cloud cost control
First, we select and upload our sample cloud platform billing data on the Data page of Kyligence Zen.
Next, we import the pre-defined YAML file to create metrics in batches in Kyligence Zen, only with one click on the Metrics page. After completing the operation, we will have the following metrics catalog.
The Metrics for Public Cloud Cost Management have been published in Kyligence Zen Metrics Template market. You may click the link and reproduce the use case in your own Kyligence Zen account.
With the above unified metrics system, users of each role in the company can select the metrics to be analyzed and the perspectives to observe the data. Take the Total Cloud Cost metric as an example. The information in the upper right corner of the metric screen shows who created the metric, the calculation method, and the analysis perspective. With this understanding in mind, the cross-departmental communication efforts will be greatly reduced, users can quickly check the cloud costs spent by each department.
Reuse metrics for data monitoring
In response to the company’s lack of budget margin tracking problem, we can create a budget control goal for cloud costs on the Goals page. We then add two sub-goals based on cloud platforms, correlate them with the metrics we created earlier, and set a budget cap, thus facilitating the management to monitor the cloud cost budget from a global perspective.
In addition, users of various roles in the company can quickly generate reports based on their business needs. For example, if an IT administrator is more concerned about cloud resource usage and daily changes, then he can generate reports on cloud resources by simply dragging and dropping, and then share them within the team.
At this point, we have completed the building of the cloud cost monitoring and analysis platform in two simple steps. For the aforementioned lag of data analysis, CSV data can be generated periodically and stored in the S3 bucket at regular intervals via ETL. Kyligence Zen will automatically retrieve CSV files with the same columns in the bucket, and treat the files as a whole to perform data queries, improving the timeliness of data analysis. (Note: We will have another blog about this topic coming soon, stay tuned!)
In response to the initial problems encountered by the company, Kyligence Zen helps to build a unified metrics store for cloud cost monitoring and analysis. Kyligence Zen’s support to incremental data upload greatly improves the analysis agility, and the dilemma of “taking stopgap measures” has become a thing of the past.
The Metrics for Public Cloud Cost Management have been published in Kyligence Zen Metrics Template market. You may click the link and reproduce the use case in your own Kyligence Zen account.
