1. Modern Data Platform: An Unbundling of a Traditional Data Warehouse
Ever wondered how modern data platforms are revolutionizing analytics?
This blog post delves into the evolution from traditional data warehouses to decentralized, modern systems. It highlights key shifts driven by cloud computing, mobile apps, and AI, and explores tools like Apache Iceberg and Spark that enhance data storage and processing. Discover the trade-offs between all-in-one solutions and custom platforms.
2. Pinot for Low-Latency Offline Table Analytics
How Uber achieves sub-second latency for its analytics?
This blog post explains Apache Pinot's offline tables enable low-latency analytics across Uber's vast data ecosystem. With Pinot, Uber handles 100+ use cases, leveraging tools like Spark and custom ingestion jobs to manage data efficiently.
https://www.uber.com/blog/pinot-for-low-latency/
3. Unpacking the Buzz around ClickHouse
Why ClickHouse is suddenly everywhere in real-time analytics?
This blog post dives into ClickHouse's speed, ease of installation, and operational simplicity, making it a favorite among developers. With its single-process architecture and scalable design, ClickHouse stands out from competitors like Apache Druid and Apache Pinot. But what are the challenges it faces, and how does it compare to emerging PostgreSQL OLAP extensions?
4. Continuous reinvention: A brief history of block storage at AWS
How AWS's Elastic Block Store (EBS) transformed from a simple block storage service to a high-performance, distributed SSD fleet?
This blog post by Marc Olson, a key architect of EBS, reveals the journey of continuous reinvention, from tackling queueing theory and noisy neighbors to leveraging custom SSD designs and advanced network protocols. Why should you read it? It offers invaluable insights into the challenges and solutions of scaling a critical cloud service, making it a must-read for anyone interested in system performance and engineering leadership.
5. Shifting E2E Testing Left at Uber
How Uber ensures seamless performance across its vast microservice architecture?
This blog post reveals Uber's journey to robust end-to-end (E2E) testing, enabling thousands of tests to run in parallel before deployment. With BITS (Backend Integration Testing Strategy), Uber achieved a 90%+ pass rate per attempt and reduced incidents by 71%. Why should you read it? It offers invaluable insights into scaling testing infrastructure and maintaining high-quality user experiences.
All rights reserved DEN Digital, India. I have provided links for informational purposes and do not suggest endorsement. All views expressed in this newsletter are my own and do not represent current, former, or future employer” opinions.