Talon.One Optimizes Cloud Infrastructure Costs

The shift to cloud computing has been one of the most significant tech trends of recent years. Moving to the cloud can bring businesses many advantages including increased agility, efficiency and security as well as the potential for innovation. Rapid scalability is another major advantage but, without the necessary controls in place, this can also lead to a rapid increase in costs. That’s why organizations today need to find ways to maximize cloud usage and efficiency.

In a recent webinar hosted by C2C, the Google Cloud Customer Community, Talon.One shared insights into how the company has optimized its cloud costs while reaping the many benefits of moving to the cloud.

Launched in 2015, Talon.One has grown into the leading enterprise-grade promotion and loyalty solution helping brands like Adidas, Afterpay and Carlsberg scale their incentive campaigns. Data is at the center of Talon.One’s powerful promotion engine, therefore monitoring the cost of data-related resources is key to ensure both performance and profitability.

Today that responsibility lies with Yotam Cohen, DevOps Team Lead. Cohen oversees Talon.One’s software development and deployment processes as well as maintaining the company’s infrastructure. “When the business started, very quickly the decision was taken to use open source PostgreSQL® as a managed service from Aiven, running on Google Cloud. Since then, Aiven has been a big part of our product. It’s been instrumental in our success and in controlling our costs as we’ve grown,” says Cohen.

Standardized plans help optimize customer environments

Each Talon.One customer has its own PostgreSQL server to guarantee security, compliance and availability. Talon.One runs more than 250 databases at any given time, totaling more than 17 terabytes in size. Aiven offers a large selection of PostgreSQL plans with standardized configurations of compute, memory and storage. This enables Talon.One to select the plan with the optimal resources for each customer’s unique requirements. With hundreds of customers, these incremental savings add up to a significant reduction in the cloud resource that Talon.One has to pay for.

Aiven has taken this one step further and created additional, customized plans to satisfy Talon.One’s needs. “The latest cost optimization exercise has reduced our PostgreSQL cost by a third, which is huge for us,” says Cohen.

“The latest cost optimization exercise has reduced our PostgreSQL cost by a third, which is huge for us,”

Scaling down as important as scaling up

Aiven for PostgreSQL® also enables Talon.One to cost-efficiently scale the databases up and down as required. “Every vendor gives you the ability to scale up performance and resources when you need them—it brings them more money after all—but scaling down can be hard when the database is already there,” says Cohen. “With Aiven, if we have a large and predictable increase in volume, we simply expand the plan and then bring it down again when everything is back to normal. It’s a refreshing approach that saves us money.”

Talon.One is also using Aiven for Apache Kafka® to minimize data pipeline costs. As the business grew, so did the amount of data ingestion, and costs began to soar. “Most companies charge you by volume so things can get out of hand quickly,” says Cohen. “By switching to Aiven for Apache Kafka, we’ve got costs under control. We have a fixed price per month per Apache Kafka® cluster which means, not only have we cut costs, the costs are predictable and will remain so as we continue to expand.”

“By switching to Aiven for Apache Kafka, we’ve got costs under control. We have a fixed price per month per Apache Kafka® cluster which means, not only have we cut costs, the costs are predictable and will remain so as we continue to expand.”

Optimizing Google Cloud costs

With customers spread across the globe, Talon.One has to consider where its resources are located to ensure its customers get the best experience possible. For example, by using the various geographical regions offered by Google Cloud, Talon.One can locate cloud resources close to its customers. It also has eight instances of Aiven for Apache Kafka to enable the company to segment customers by geography and, in addition to having a dedicated PostgreSQL server, each customer is allocated to a regional Kafka cluster.

Optimizing Google Cloud costs is a key part of the organization’s cloud cost management strategy. “Google Cloud has a great dashboard for billings which provides an aggregated view across many projects. If you label your infrastructure properly, you can break down the costs really well and find opportunities for cost optimization. We are very happy that Aiven services are hosted on Google—it helps us keep on top of our cloud costs,” says Cohen.

Having PostgreSQL and Kafka managed by Aiven is another way for Talon.One to optimize resources. “We’d need at least two or three full time employees to do the work that Aiven does but we wouldn’t have the same quality of service,” says Cohen. “We take it for granted that everything runs seamlessly with Aiven. Upgrades take place with no interruptions. We touch PostgreSQL maybe once a month.”

“We’d need at least two or three full time employees to do the work that Aiven does but we wouldn’t have the same quality of service,” says Cohen. “We take it for granted that everything runs seamlessly with Aiven. Upgrades take place with no interruptions. We touch PostgreSQL maybe once a month.”

Seamless integration avoids more costs

When onboarding new tools, Cohen’s team invests money, time and effort in researching how to integrate them with the existing ecosystem. For Talon.One, this additional cost is avoided with Aiven. For example, when the team started using Aiven for Apache Kafka Connect and Aiven for ClickHouse®, the new technologies integrated seamlessly with PostgreSQL and Kafka.

Looking to the future, Cohen anticipates more growth, particularly as the company increasingly attracts larger enterprises. “We’re becoming even more data-oriented which means more data ingestion and more analytics. Visibility into, and management of, data costs will remain very important. Aiven is our go-to partner for anything data related and will remain key to our cost optimization efforts as we journey into the future,” Cohen concludes.

To get the latest news about Aiven and our services, plus a bit of extra around all things open source, subscribe to our monthly newsletter! Daily news about Aiven is available on our LinkedIn and Twitter feeds.

If you just want to find out about our service updates, follow our changelog.