Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

This case study is about an engagement which had an enterprise-wide data solution. This was an Azure Data Warehouse PaaS Solution and the objective here was to measure the carbon savings brought in due to the optimizations done by the team. It was a complete enterprise-wide data solution that is a centralized data platform that was being built across all the streams. As part of it they also have BI, Machine Learning and Data Analytics. Their primary challenge here was that their spend on cloud was very high and they wanted to reduce it. So, an analysis was done, and it was determined that the Queries were not very well written and needed lot of optimizations without which lot of computes was getting wasted. Further, Query optimizations were done and the spend on cloud also came down.To measure the carbon savings associated with this, the logic applied was to determine the number of Virtual Machines that would have come down due to the optimizations and use that as a factor to measure. Azure Data Warehouse is PaaS Solution and is a cluster of VM’s. Here the concept is that as the number of ADW units increases the VM also increases and if ADW units decrease the number of VM also decreases. So, the client typically requests Microsoft on the required number of ADW units and would pay the cost associated with it. As the ADW units are allocated, internally the VMs get allocated. However, the number of VMs are not visible to us. The data points available for calculation was Azure Data Warehouse units allocated and the spend on cloud in Pre-Optimization and Post-Optimization state. As a next step to determine the total number of VM’s, the cost of a single VM was determined through Microsoft pricing calculator considering the instance details of vCPU, RAM etc. Considering the Total cost of all VM’s and the single cost of a VM, the total number of VM’s allocated were derived. These details along with the CPU Utilization %, location of the servers were used to calculate the carbon savings using Client Carbon Impact Calculator tool, in which we entered the data for the pre and post scenarios to calculate the emission savings. Furthermore, we assessed the delivery impact of the service offered to the client and deducted it from the carbon savings obtained by reducing the number of virtual machines. Also, as part of our learning journey towards sustainability, please review the Key Takeaways from the Case Study, which will assist you in presenting this as a plus point to the client.