Tonkean Simplifies Database Management, Achieves 99.99% Uptime on AWS

Tonkean, a San Francisco–based software-as-a-service startup dubbed the “operating system for business operations,” offers companies a no-code, enterprise-grade solution that simplifies the orchestration of complex business processes. To operate this solution, it needed a reliable way to run, manage, and scale open-source MySQL. Built entirely on the cloud, Tonkean was experiencing problems with its original cloud service provider. The service was difficult to manage and required too much time from its small development team, which was busy safely scaling the database, protecting the network, and adding new product features. 

Tonkean was using Redis for caching to improve performance, but it would frequently crash when it reached peak load. Recovering Redis wasn’t simple because the cloud service’s standby server wouldn’t fail over, pausing updates to the database and causing data loss. So the availability and reliability of Tonkean’s product suffered. “Our customers trust us to be available. If our database is down, Tonkean cannot run—it’s the heart of our company,” says Afik Udi, senior manager of production and infrastructure at Tonkean.

In 2019, Tonkean migrated to Amazon Web Services (AWS) and began using Amazon Relational Database Service (Amazon RDS) for MySQL, which enabled it to set up, operate, and scale its database on the cloud. Migrating to a managed service saved its development team time and resources so it could focus on innovation. The migration also increased the reliability and availability of Tonkean’s product. By using Amazon RDS Read Replicas—which enable engineers to create replicas of database instances and serve high-volume application read traffic from multiple copies of data—Tonkean can elastically scale beyond the capacity constraints of a single database instance for read-heavy workloads. The migration drastically improved Tonkean’s support for its customers by reducing the risk of downtime and data loss.

Sagi Eliyahu, CEO of Tonkean, and Offir Talmor, chief technology officer, founded the company in 2015 to address a growing need for a flexible workflow automation service that didn’t require writing and maintaining custom scripts. Since then, the startup has tripled its head count to more than 50 employees and opened two offices, one in San Francisco and the other in Tel Aviv. Because Tonkean was acquiring new enterprise customers so quickly—attracting the attention of TripActions, EVERSANA, Hopper, and others—it needed a solution to address reliability, availability, and scalability issues.

Tonkean chose to migrate to a cloud service that would lessen the burden of management on its developers. The development team decided to use Amazon RDS for MySQL, enabling it to manage MySQL deployments in the cloud and quickly deploy MySQL servers with cost efficient and resizable hardware capacity. The managed service would also provide high quality and accessible support documentation and direct support from AWS team experts. And to assist with the quick and safe scalability required, Amazon RDS would enable it to launch read replicas, a feature absent from its previous cloud solution.

Tonkean migrated to Amazon RDS for MySQL in 2019. With support from AWS, Tonkean completed the migration in just 8 weeks, and the process was smooth and painless. Once the development team connected to the database in Amazon RDS, downtime became virtually nonexistent. The team uses Amazon RDS as its main data store, which currently houses 750 GB of data. “All the data that drives the engine at Tonkean is inside Amazon RDS,” explains Udi. “We rely heavily on the performance of our database and the stability of the service housing it. And we’re fully confident in AWS.”  

Tonkean also uses Amazon ElastiCache, which enables developers to seamlessly set up, run, and scale popular open-source-compatible in-memory data stores in the cloud. Using Amazon ElastiCache for Redis, which combines the speed, simplicity, and versatility of open source Redis with manageability, security, and scalability from AWS, Tonkean can achieve the stability and reliability it needs. “Since the migration, Redis has performed better than in our previous cloud environment and hasn’t lost data,” says Udi. To run Tonkean’s services, developers are using a managed Kubernetes cluster on Amazon Elastic Kubernetes Service (Amazon EKS). Amazon EKS gives developers the flexibility to start, run, and scale Kubernetes applications in the AWS Cloud or on premises, and it supports Tonkean by freeing its small development team from managing and deploying Kubernetes servers.  

Since migrating to Amazon RDS, Tonkean has seen an increase in availability. The uptime of its data integrations increased from a 99.76 percent average before the move to 99.94 percent after. The uptime of its web applications increased from 99.96 percent to 99.99 percent. Tonkean can also safely and quickly scale as needed, using Amazon RDS Read Replicas and Amazon Elastic Compute Cloud (Amazon EC2), a web service that provides secure and resizable compute capacity in the cloud. By using Amazon EC2 Spot Instances, the company can take advantage of unused Amazon EC2 capacity at up to a 90 percent discount to rapidly scale up when necessary. Scalability became notably important when some Tonkean customers faced disruptions during the COVID-19 pandemic. “We had a travel-related company receiving thousands of support tickets when customers began canceling travel,” Udi recounts. “We helped support that sharp increase in traffic, and doing it without the scaling capabilities on AWS would have been far more challenging.” 

Tonkean’s production environment is operating at its largest size to date, peaking at 9 million rows per second and 120 million requests daily to Amazon RDS. Yet the company has managed to save 15 percent on costs since the move. Because Tonkean is devoting limited resources to running the infrastructure, self-managing the database isn’t feasible. And now that Tonkean is using managed services on AWS, self-managing isn’t necessary.

Tonkean plans to continue optimizing its product and database on AWS, and it may develop several smaller single-tenant environments modeled after its production environment to offer customers dedicated instances of Tonkean: many of its enterprise customers require data that is separate from that of other customers and secure ways to connect their on-premises data sources to enhance internal automation workflows. 

Because of the migration to Amazon RDS, Tonkean can offer its customers a reliable, available product with support from a database that can scale quickly, safely, and cost effectively in a stable, fully managed cloud environment. Its small developer team no longer needs to devote time to database management, and that makes all the difference. “I think in many respects, Amazon RDS helps us sleep better at night,” says Talmor.