Architecting Agentic Data Planes (Lessons from OpenAI's billion-dollar Postgres mistake)

👉 Agentic AI systems place new demands on data infrastructure: high write concurrency, bursty traffic, and mixed transactional retrieval workloads. This session uses OpenAI’s recent PostgreSQL scaling challenges as a case study to analyze where traditional database architectures break down, examining the operational trade offs and hidden costs that emerge when legacy systems reach their limits. Matthew Penaroza (Head of Solutions Architecture at SurrealDB) demonstrates how to solve these bottlenecks by moving toward a natively distributed data plane, focusing on implementing horizontal write scalability, utilizing multi model access (graph, document, and vector), and serving both transactional and AI workloads from a single live dataset to avoid the architectural "duct tape" of traditional databases.

🖥️ Explore other SurrealDB webinars: https://sdb.li/4sOML5J
📃 Read the blog: https://sdb.li/4mLqtjV

Chapters:

00:00:00 Intro
00:00:10 Introduction and Agenda
00:01:26 Speaker Background and Credibility
00:03:05 What Is an Agentic Data Plane?
00:05:56 OpenAI at Scale: Context
00:07:15 OpenAI's Architecture: Postgres + Replicas
00:14:04 Core Constraints: Single-Writer Bottleneck
00:16:43 The Cost of 50 Read Replicas
00:21:11 Write Pressure → Complexity and Containment
00:28:08 Agentic Workload Requirements
00:34:02 A Distributed Alternative (SurrealDB)
00:47:09 Q&A

--- Resources ---
☁️ Start free on Surreal Cloud: https://surrealdb.com/cloud
🎓 SurrealDB University: https://surrealdb.com/learn
📚 Docs: https://surrealdb.com/docs
🎩 Surrealist (GUI): https://surrealdb.com/surrealist
⭐️ GitHub: https://github.com/surrealdb/surrealdb

--- What is SurrealDB? ---
SurrealDB is an AI-native, multi-model database built in Rust that unifies document, graph, time-series, relational, geospatial and key-value data models into a single engine. Combined with powerful search and retrieval (vector, full-text, hybrid), real-time and event-driven capabilities, and Spectron - its persistent memory layer for AI agents - SurrealDB provides the complete context layer so that models run next to the data, context remains synchronised, and application logic stays simple and dependable.

--- Connect ---
🐦 Twitter: https://twitter.com/surrealdb
👩‍💻 LinkedIn: https://www.linkedin.com/company/surrealdb/
💬 Discord: https://discord.com/invite/surrealdb
🛍️ Merch: https://surrealdb.store/

Видео Architecting Agentic Data Planes (Lessons from OpenAI's billion-dollar Postgres mistake) канала SurrealDB