Saving $2.2 million annually by modernizing on Amazon Aurora with Infios

Infios provides adaptable supply chain solutions to businesses of almost all sizes, helping them simplify operations, optimize efficiency, and drive measurable impact in their fields. The company serves brokers, shippers, and carriers in 70 countries worldwide.

Infios had deployed its workloads in Oracle but needed to reduce the cost of technical support and the price of running multiple clusters. Further, Infios wanted to avoid the increased licensing costs of renewing its contract with its database provider. So, the company began to explore other options as it approached the end of its service agreement. “We looked at ways to more effectively use our data to support specific use cases,” says Adam Jones, senior vice president of cloud engineering at Infios.

The company also saw the opportunity to improve its overall data footprint and comply with the requirements of the American National Standards Institute (ANSI), deciding to migrate to Aurora PostgreSQL-Compatible Edition. “The more we dug into it, the more it seemed like a no-brainer to use Aurora instead of trying to reinvent the wheel ourselves,” says Jones.

Additionally, Infios expanded its DR capabilities by using Aurora global clusters, which are designed for globally distributed applications, allowing a single Aurora database to span multiple AWS Regions. This way, Infios can automatically replicate its databases to a secondary instance and perform failover to another Region as needed. “The management features on top of Aurora make it simple to figure out how we consume storage and use resources, which keeps our costs predictable and low,” says Jones.

The Infios team refactored its Oracle software code into ANSI SQL for interoperability with Aurora. Then, the company accelerated its migration by using AWS Database Migration Service (AWS DMS), which has been trusted by customers to securely migrate over 1.5 million databases with minimal downtime.

The company also engaged AWS Enterprise Support to achieve its desired business outcomes on AWS. “When you’re rapidly modernizing, there’s always going to be gaps in what you don’t know, especially for smaller companies that might not be able to hire a principal engineer,” says Nicholas Mariani, senior director of cloud engineering at Infios. “By collaborating with the AWS team, we can close the gaps in our knowledge and optimize our solutions.”

In 6 months, Infios migrated more than 200 databases to Aurora, each of which supports one or more customer tenants. To minimize downtime, the Infios team used the monitoring capabilities in Aurora to analyze its cluster performance. “Having these capabilities was super helpful to debug performance issues,” says Jones.

By migrating to Aurora, Infios reduced the volume of data it stores in production. “Because we’re charged only for the storage that we actually consume on Aurora, we cut the amount of data that we’re paying for in half,” says Jones. Infios decreased its data footprint from 128 to 60 TB in Aurora. The company also adopted Aurora I/O-Optimized, a configuration that provides improved price performance and predictable pricing for I/O-intensive workloads.

Using Aurora, Infios saves USD $1.2 million on database costs annually. The company also saves another USD $1 million on licensing costs, bringing its total annual cost savings to USD $2.2 million. “It’s been different to work with a cloud provider that helps us lower our costs,” says Kelly Harrelson, database administration manager at Infios.

Migrating to Aurora has transformed Infios’s developer workflows, freeing its teams to focus on high-value projects. Database refreshes for multiple environments now take as little as 1–2 hours, instead of more than 24 hours. “We used to spend about 95 percent of our time on troubleshooting performance issues, managing backups, and creating copies for lower environments,” says Harrelson. “Because Aurora storage is self-healing, those tasks are now a very minimal part of our job.”

Developers also benefit from using Aurora cloning to quickly and cost-effectively create copy-on-write clones of their databases. Database replication now takes under 20 minutes, instead of up to 3 days previously.

Using Aurora, Infios has also reduced database timeouts by 45 percent, delivering a more reliable service to customers. “Now that each tenant has its own dedicated compute and storage, customers experience a much more stable, predictable performance,” says Jones. “We can also fine-tune performance on a tenant-by-tenant basis.”

The company will continue to boost developer productivity and infrastructure stability by standardizing its cloud environments on Kubernetes. As part of these efforts, Infios will migrate its application layer to Amazon Elastic Kubernetes Service (Amazon EKS) so that teams can start, run, and scale Kubernetes without thinking about cluster management.

“We’ve worked with the AWS team to identify where to focus our efforts and how to implement best practices,” says Mariani. “Now that we’ve modernized the biggest pieces of our infrastructure, we’re going to push all our solutions in that direction.”