Mydbops’ Post

Ray Paik and Daniël van Eeden sat down with Airton Lastori (Product Manager at ‪TiDB, powered by PingCAP‬ to discuss how #MySQL users can evaluate whether #TiDB is the right #database for them and the tools available to help with migration on the latest episode of Data In The Hallway. If you’re using MySQL and open to learning about alternatives for scale, reliability, and architectural coherence, this one’s for you. https://lnkd.in/gkcBAV-2

Preload extensions once, use them everywhere. Set shared_preload_libraries at the branch level and every database on that branch shares the same foundation. Example: pg_stat_statements, pg_trgm → consistent query stats + text ops across envs. Notes: applies to all DBs on the branch; enabling/changing preloads triggers a restart. Add your list in Branch Settings → Preload libraries.

just discovered storage attached indexing in Cassandra before SAI Cassandra secondary indexes were slow and inefficient key issues 1. only worked well for simple "equals" queries 2. slowed down badly with more than one index 3. used too much memory 4. forced every query to check all nodes even if it did not need to this made it hard to search or filter data flexibly without redesigning your whole table. SAI fixed this by storing indexes smarter right next to the data so queries are faster, use less memory and actually scale

🎓 System Design Bangla Series | Episode 2.3: Database Replication This episode focuses on one of the pillars of scalable architecture — Database Replication. We discuss: ✅ How replication enhances fault tolerance & performance ✅ Different replication models (Master–Slave, Multi-Master) ✅ Trade-offs between consistency and availability ✅ Real-world use cases in large-scale systems If you’re a backend developer or preparing for system design interviews, this concept is a must-know! 🎥 Watch the full episode here → https://lnkd.in/greZkkJf #SystemDesign #DatabaseEngineering #HighAvailability #BanglaTutorial #BackendDevelopment

Why do we even need tools like pgbouncer or pgcat for Postgres? And what does transaction pooling even mean? Let's play around with transaction pooling in pgcat to see.

Creating Embeddings for RAG Systems In a RAG pipeline, once chunks of documents are ready, they need to be embedded — converted into vectors that capture their semantic meaning. Oracle 23ai lets you generate these embeddings using models inside or outside the database. You can even import ONNX-format embedding models directly into Oracle 23ai to keep all processing within the database, ensuring data security and speed.

Devlog #3 implemented metadata, it also gets updated if new incoming request with different data. Empty metadata will be ignored and the value from db will be returned. If you want to delete metadata provide null as value for it. Metadata is always returned even if null view on x: https://lnkd.in/dGUSFYtP

Why “It’s Just a File” Is the Most Dangerous Assumption in Data Migration When teams think about migration, they often plan around systems, tables, and connections. But one silent saboteur often slips through every checkpoint: file formats — especially the humble CSV. https://lnkd.in/gVuTihud

I highly recommend this paper for anyone interested into databases or just want to start with reading whitepapers - "Bitcask: A Log-Structured Hash Table for Fast Key/Value Data". Read this paper if you want to build your own database and learn concepts which powers modern tech like Kafka, LSM-trees and Write Ahead Log. It shows the power of append-only log and sequential writes.

Finding the most common value in a dataset just got easier. The new mode() function in #QuestDB lets you identify the most frequent value in a column with a simple SQL statement: SELECT symbol, mode(price) FROM trades WHERE timestamp IN today();