Maximizing Efficiency for Bidding System Using AWS Graviton–Based Instances with Jampp

Founded in 2013, Jampp is a programmatic advertising provider that helps businesses drive growth, whether they want to reach new users or increase conversions. In 2021, it was acquired by Affle, a global consumer intelligence technology company. Jampp uses machine learning to analyze user behaviors and demographics, predict user actions and interests, and automate and optimize campaigns, as well as retarget users who are most likely to reengage.

Jampp requires compute power that is not only reliable but also exceptionally fast. Transactions must be completed in milliseconds, and the data load can fluctuate to millions of transactions per second. To support this environment, Jampp uses thousands of Amazon Elastic Compute Cloud (Amazon EC2) instances, which offer secure and resizable compute capacity for virtually any workload. The company has used AWS since its founding.

“Why did we choose AWS over other competitors? It came down to compliance aspects, such as disaster recovery and the General Data Protection Regulation, as well as the ability to operate in different data centers without significant latency. We found AWS to be the most suitable for these needs,” says Charles Yong, chief architect and technology officer at Affle. “AWS might not have had all the capabilities it has now, but it has continually adapted to meet our needs.”

To improve efficiency without compromising performance, Jampp adopted AWS Graviton processors, which are designed to deliver excellent price performance for cloud workloads running on Amazon EC2.

When Jampp began to migrate to AWS Graviton–based instances, its DevOps team ran both its existing instances and AWS Graviton–based ones in parallel, which provided a reliable fallback option. The team first migrated non-production workloads, and then gradually migrated higher-impact production workloads. This strategy resulted in a smooth migration with no noticeable impact on users.

Jampp uses Amazon EC2 Spot Instances—which help users take advantage of unused Amazon EC2 capacity—to dynamically scale its infrastructure up and down depending on the availability of AWS Graviton–based instances.

Jampp also employs Amazon EMR, a cloud big data solution for petabyte-scale data processing, to deploy Apache Spark and transfer data across AWS Regions and Availability Zones. This helps the company perform real-time bidding processes at scale and maintain high availability. Additionally, the company uses Amazon Elastic Container Service (Amazon ECS), a fully managed container orchestration service, to deploy containerized services using AWS Graviton–based instances, which further enhances availability.

By incorporating AWS Graviton–based instances into its bidding system, Jampp unlocked a 10 percent improvement in efficiency, particularly in instance throughput. Jampp can win bids faster and improve its win rate, meeting the needs of data-savvy and sophisticated advertisers and expanding its reach.

“To be the go-to solution that mobile advertisers use to market at scale, we need to continually explore innovative ways to enhance our services,” says Yong. “AWS Graviton–based instances have played a pivotal role in advancing this objective by significantly improving our performance.”