How Uber Uses Kubernetes

Uber Engineering details their migration from a legacy monolithic monitoring system to a modern, cloud-native observability platform for their corporate network infrastructure.

This article discusses the implementation of zone failure resilience in Apache Pinot at Uber, detailing strategies to ensure uninterrupted service during zone failures.

The article announces that the Cadence project has joined the Cloud Native Computing Foundation (CNCF), highlighting its commitment to open-source development.

Uber's migration from Spark 2. 4 to Spark 3. 3 involved upgrading over 2 million Spark applications, utilizing innovative automation tools like Iron Dome.

The article discusses Uber's implementation of encryption at rest and disk isolation at scale using their Stateful Platform, Odin.

The article discusses Uber's automation of Kerberos keytab rotation, detailing the challenges faced and the solutions implemented through their Keytab Distribution Pipeline (KDP).

This article discusses the development of Uber's Multi-Cloud Secrets Management Platform, designed to enhance security across its extensive microservices architecture.

The article discusses Uber's migration of large-scale interactive compute workloads from Peloton to Kubernetes, focusing on minimizing disruption while enhancing resource management and cloud readi...

The article discusses Uber's migration from Apache Mesos to Kubernetes, detailing the motivations, challenges, and solutions encountered during the transition.

This article discusses Uber's implementation of elastic resource management on Kubernetes, focusing on enhancements made to support Ray-based job management.

Uber's blog post discusses their migration of machine learning workloads to Kubernetes using Ray, detailing the challenges faced with their previous setup and the improvements achieved with the new...

The article discusses the MySQL fleet at Uber, which consists of over 2,300 independent clusters that support critical operations for the platform.

The article discusses how Uber utilizes Ray®, a general compute engine for Python®, to enhance the efficiency of its rides business through improved machine learning model performance and optimizat...

The article discusses The Accounter, a global coordination system developed by Uber to enhance operational throughput and safety on its stateful platform, Odin.

The article discusses how Uber optimizes the training of Large Language Models (LLMs) using both open-source and in-house models.

Odin is Uber's stateful platform designed to manage various technologies for data storage efficiently.

The article discusses Uber's evolution in machine learning (ML) through its centralized platform, Michelangelo, highlighting its transition from predictive to generative AI.

The article discusses Uber's journey in scaling its AI/ML infrastructure, highlighting the transition from on-premise to cloud solutions, the implementation of new technologies, and the optimizatio...

The article discusses Uber's journey in enhancing its Palette Meta Store, focusing on the challenges faced, the solutions implemented, and the resulting improvements in machine learning feature man...

The article discusses Uber's experience in adopting SPIFFE/SPIRE at scale, detailing the challenges and solutions encountered in implementing a Zero Trust security model across a complex microservi...

The article discusses Uber's transition to a portable microservices architecture optimized for cloud deployment.

This article discusses Uber's journey in adopting arm64 architecture using the Zig programming language for their infrastructure.

The article discusses how Uber has enhanced developer productivity through the implementation of DevPods, a remote development environment.

The article discusses Uber's implementation of an automated vertical CPU scaling system that optimizes resource allocation for storage workloads, leading to significant cost savings and improved re...

Uber's article discusses the implementation of a highly scalable and distributed Remote Shuffle Service (RSS) designed to enhance the efficiency of data processing in Apache Spark.

The article discusses strategies to avoid CPU throttling in a containerized environment, particularly at Uber, where stateful workloads run on a large fleet of hosts.

This article discusses how Uber's engineering team optimized garbage collection (GC) in their Go-based services, resulting in a significant reduction of compute costs amounting to 70,000 cores save...

The article discusses Uber's approach to managing compute clusters efficiently and reliably at scale, emphasizing the importance of CPU utilization and resource scheduling.

The article discusses the integration of Elastic Horovod with Ray, focusing on how this combination enhances distributed deep learning training by enabling autoscaling and fault tolerance.

The article discusses Fiber, a distributed computing library developed by Uber that simplifies large-scale computation for AI and machine learning applications.

The article discusses Uber ATG's machine learning infrastructure and versioning control platform, VerCD, designed to manage the complexities of developing self-driving vehicles.

The article discusses Uber's open source initiatives in 2019, highlighting the company's contributions to the open source community, the establishment of the Open Source Program Office (OSPO), and ...

The article features an interview with Felix Cheung, Data Platform Engineering Manager at Uber, discussing the advantages of open source software in private enterprise.

The article features an interview with Celina Ward, an M3 Observability Engineer at Uber, discussing her journey into engineering, the open-source M3 metrics platform, and her experiences presentin...

The article reviews Uber's open source initiatives in 2018, highlighting the diversity of projects and their impact on the technical community.

The article introduces Makisu, Uber's open-source Docker image builder designed for Apache Mesos and Kubernetes.

The article discusses Peloton, Uber's unified resource scheduler designed to manage diverse cluster workloads efficiently.

The article highlights seven open source projects showcased at the Uber Open Summit, emphasizing Uber's contributions to the open source community.

The article features an interview with Yuri Shkuro, the Jaeger Project Lead at Uber, discussing the development and open sourcing of Jaeger, a distributed tracing system.