Skip to main content
Engineering, Backend

The Accounter: Scaling Operational Throughput on Uber’s Stateful Platform

November 21 / Global
Featured image for The Accounter: Scaling Operational Throughput on Uber’s Stateful Platform
Image
Figure 1: Example of conflicting operations resulting in cluster availability loss.
Image
Figure 2: Overview of the cluster health collection on Odin.
Image
Figure 3: Overview of the data model used to represent the relationship between operations and groups.
Image
Figure 4: The Accounter architecture and claim granting process.
Image
Figure 5: Timeline of Uber’s transition to encryption-at-rest.
Jesper Borlum

Jesper Borlum

Jesper Borlum, Sr. Staff Engineer at Uber, is a seasoned software engineer, architect, and team player. He leads the Stateful Platform team, responsible for building the infrastructure to manage all of Uber’s stateful systems. The team’s mission is to deliver a fully self-healing platform without compromising availability, reliability, or cost. He’s currently leading the effort to adopt Arm at Uber.

Gianluca Mezzetti

Gianluca Mezzetti

Gianluca Mezzetti, Sr. Staff Engineer at Uber, was among the pioneers of the Stateful Platform team. His extensive contributions across multiple platform domains, such as workflows, concurrency control, host remediation, goal state storage, and auditing, have been instrumental in expanding the platform’s capacity. Currently, he leads the initiative to integrate Kubernetes into Odin.

Alexander Blazhenskikh

Alexander Blazhenskikh

Alexander Blazhenskikh, Sr. Software Engineer at Uber, is a member of the Stateful Platform team. He contributes to the Accounter, a critical concurrency control service, addressing safety, consistency, and scalability challenges with expertise and precision.

Posted by Jesper Borlum, Gianluca Mezzetti, Alexander Blazhenskikh