Uber has one of the largest Apache HadoopⓇ Distributed File System (HDFS) deployments in the world, with exabytes of data across tens of clusters. HDFS team at Uber had to solve the problem of…

With the introduction of Model Excellence Scores at Uber, we're setting a new standard for measuring, monitoring, and maintaining ML model quality–read how this innovative approach aims to enhance ML…

Machine Learning (ML) is celebrating its 8th year at Uber since we first started using complex rule-based machine learning models for driver-rider matching and pricing teams in 2016. Since then, our…

Uber connects the physical and digital worlds to help make movement happen at the tap of a button. Billions of trips, deliveries, and tens of billions of financial transactions across earners,…

This week, we'll dive into how we migrated Uber’s business-critical ledger data to LedgerStore. We'll detail how we moved more than a trillion entries (making up a few petabytes of data) transparently…

CheckEnv是Uber开发的解决微服务架构中复杂调用链和跨环境通信问题的工具。它利用Grail和Local Graph平台监控和解决跨环境的RPC调用，并提供API进行异常检测、故障排除和优化。CheckEnv是Ballast负载测试平台的首个用户，通过MazeX数据摄取管道更新微服务架构的依赖关系。未来，CheckEnv将扩展数据摄取能力，构建更强大的图形，提升微服务的可靠性和效率。通过图形化分析，CheckEnv能够优化数据流，提高服务的整体效率。它可以应对微服务架构中的复杂挑战，如实时故障检测、事件预测、异常检测、效率改进和工作流管理等。

uVitals是一个用于异常检测的系统，提供了多种模型和功能。它的后端服务使用Go编程语言构建，通过CRUD操作可以有效地管理指标配置。系统核心是“异常”API，可以提供给定指标和日期范围内的异常列表。用户友好的界面使用户可以轻松地导航和自主操作。uVitals不仅能够检测异常，还能帮助用户理解异常，并提供详细的数据可视化和深入分析功能。系统还提供及时的异常通知和用户反馈机制，以提高准确性。通过分析数据的分布和特征，uVitals可以减少低流量城市的误报。此外，uVitals还在性能方面进行了优化，采用Apache Spark®替代Pandas®，平均提高了77%的运行效率。未来，uVitals还将进一步发展，包括实时异常检测和利用AI提供更丰富的内容。

Uber利用微服务和基于属性的访问控制模型来实现灵活的授权策略。它使用授权引擎和属性存储来评估条件表达式和获取属性值，并选择了Google的CEL作为表达式语言。Uber结合基于属性的访问控制（ABAC）和基于角色的访问控制（RBAC）来满足不同的授权需求，并使用uOwn服务的所有权信息创建通用策略。策略管理中心Charter简化了策略的管理和分发。通过引入属性存储和CEL语言，Uber提供了更大的灵活性和可扩展性，提高了授权效率，节省了工程时间，并降低了系统的复杂性。总之，Uber通过采用属性访问控制策略，为所有服务提供了高效且可扩展的授权框架。该策略有助于保护Uber的数据安全和用户隐私。

With the evolution of storage table formats Apache Hudi®, Apache Iceberg®, and Delta Lake™, more and more companies are building up their lakehouse on top of these formats for many use cases, like incremental ingestion. But the speed of upserts sometimes is still a problem when the data volumes go up.

In storage tables, Apache Parquet is used as the main file format. In this article, we will discuss how we built a row-level secondary index and the innovations we introduced in Apache Parquet to speed up the upsert data inside a Parquet file. We will also demonstrate benchmarking results that show much faster speeds than traditional copy-on-write in Delta Lake and Hudi.

Uber's Financial Computation Service processes various types of money movements from multiple upstreams, generating journal entries that can be linked to actual transactions for accuracy. A rulePlan, a directed tree JSON representation, guides execution. Audit events are created for each transaction, containing all information needed to trace back to business flow and rules. Recompute and reconciliation are possible to ensure reproducibility. All upstream events are consumed and processed by the service.

Uber has made a public commitment to phase out carbon emissions in the United States, Canada, and Europe by 2030, and worldwide by 2040. We maintain periodic updates on our progress via our Climate Assessment and Performance Report, which shows both how far we’ve come and how far we have yet to go.

Underlying this report is a large effort to prepare the data presented within, everything from identifying vehicle fuel type across markets, to carbon emissions per vehicle-mile and ultimately to passenger-mile traveled. In this blog post, we’ll take a closer look at the complexities of identifying “green” vehicles onboarded to Uber, and our solutions for managing those data.

Apache Spark™ is a widely used open source distributed computing engine. It is one of the main components of Uber’s data stack.

Spark is the primary batch compute engine at Uber. Like any other framework, Spark comes with its own set of tradeoffs.

Uber has one of the largest Hadoop® Distributed File System (HDFS) deployments in the world, with exabytes of data across tens of clusters. It is important, but also challenging, to keep scaling our data infrastructure with the balance between efficiency, service reliability, and high performance. As a cost efficiency improvement effort that will save us tens of millions dollars every year, we aim to adopt higher density HDD (16+TB) SKUs to replace existing SKUs with 4TB HDDs that are still used by the majority of our HDFS clusters.

One of the biggest challenges when fully adopting high-density disk SKU comes from the disk IO bandwidth. While the capacity of each HDD increases by 2x to 4x, the I/O bandwidth of each HDD does not increase accordingly. This may cause IO throttling when DataNodes serve read/write requests. This can be seen from the chart below, which shows the trend of slow read packet read count from one DataNode. Given the persistent and sizable number of slow read occurrences, it is important to find new approaches to prevent performance degradation.

All the best things come in threes: the Three Musketeers, the Three Stooges, and, of course, your favorite three-cheese pizza ordered via the UberEats app. Engineering Security (EngSec) at Uber agrees and we have formed our own trio for how we simulate cybersecurity incidents at Uber to exercise our ability to act decisively should an incident occur. This three-pronged approach consists of tabletop exercises, red team operations, and atomic simulations.

In November 2021 we decided to evaluate arm64 for Uber. Most of our services are written in either Go or Java, but our build systems only supported compiling to x86_64. Today, thanks to Open Source collaboration, Uber has a system-independent (hermetic) build toolchain that seamlessly powers multiple architectures. We used this toolchain to bootstrap our arm64 hosts. This post is a story with how we went about it, our early thinking, problems, some achievements, and next steps.

At Uber, we obsess over delivering highly performant and reliable experiences to our partners and customers. We treat degradations to app performance the same way as any other functional regressions.…