Visualizing cloud spend across the organization with the DoiT console
Bdeo began working with DoiT in 2023, leveraging DoiT Cloud Analytics and building custom dashboards to gain clearer insight into its costs and how to manage them.
“Our DevOps engineers, our ML Ops, our CIO – every role related to our infrastructure now has a tool to visualize any cost in detail within five minutes,” Fernandez Reyes explains. “If we want to understand why costs are rising or which exact resource is costing a certain amount, we just open the report, configure it, and in five minutes we have the answer.”
Bdeo now uses Attributions in the DoiT Platform to attribute costs to specific products and services and combine them with others to get a clear understanding of complex costs throughout its cloud services. By visualizing its expenditure in this way, Bdeo can then identify patterns of costs in various parts of the business, helping the company understand where efficiencies can be made. The console has even helped Bdeo identify costs it didn’t know it had, such as an encryption service that it didn’t need, leading to quick fixes to eliminate unnecessary expenses.
Previously, Bdeo also used to experience a high rate of false positive alerts on cost anomalies, incorrectly identifying unexpected price surges. Now, using DoiT Anomaly Detection, it can correctly identify unintentional cost spikes. Bdeo no longer experiences erroneous alerts and is instead able to quickly identify surging costs. Anomaly Detection also gives Bdeo recommendations for how to reduce these anomalous costs, helping keep expenses under control.
Cutting the cost of on-demand instances with Flexsave
Much of Bdeo’s EC2 workloads are covered by AWS Reserved Instances, giving it discounts for long-term usage commitments. However, Bdeo needs to be able to turn a portion of its machines on and off at certain times, meaning it is unable to commit to Reserved Instances for all its workloads, where it uses on-demand pricing instead.
To help Bdeo optimize its costs for these on-demand workloads, DoiT recommended Flexsave, which automatically identifies and applies discounted rates to any of Bdeo’s EC2 workloads that are not already covered by commitments, giving Bdeo a 10% discount on all its on-demand costs.
Working together for cost-efficient architecture
Bdeo has also been working with DoiT Cloud Reliability Engineers (CREs) to identify more efficiencies that can be made across its cloud infrastructure. For example, with so much data dedicated to images and video, Bdeo wanted to find the most efficient way to tier its data storage without compromising on quality. It was a natural partnership given DoiT’s expert extensive knowledge thanks to their close relationship with AWS. DoiT CREs worked closely with the Bdeo team to come up with an intelligent tiering system to enable it to structure its storage pricing to give it quick access to the files it needed for the optimum price.
For Fernandez Reyes, it was a good example of the collaborative relationship between Bdeo and DoiT. “We have a really technical team, with strong tech skills, so when we face something that we don’t understand, or that is going to be very time consuming to work on, we need to have a very technical partner. DoiT has highly knowledgeable engineers, they always respond within two to three hours, and we can go as deep as we want into any problem or any issue.”
Optimizing infrastructure with the Well-Architected Framework review
As Bdeo gained a clearer understanding of its costs and worked to reduce them, it also began the process of analyzing and optimizing its cloud architecture for reliability, performance, and security. Together with DoiT CREs, Bdeo worked through a Well-Architected Framework review of its AWS infrastructure to understand where it could be improved.
“We spent six to eight hours with DoiT, reviewing all the pillars and answering the questions,” Fernandez Reyes recalls. “We didn’t want to just pass the exam, or to have the certificate to show we have great architecture, we wanted to see where we could improve. It was really useful to work with DoiT on this, and we learned a lot about our infrastructure in each pillar.”
Following the review, Bdeo immediately implemented a number of security improvements to reduce its systems’ vulnerability. The review also helped Bdeo focus on the health and performance of individual workloads, such as the speed of its AI processing, to ensure it is able to continue to meet certain performance metrics as it scales up its infrastructure.
