proomt

Search

Search posts, papers, and topics

Hall of Fame

Hall of FameJoseph M. Hellerstein, Michael Stonebraker, James Hamilton2007161 min readpaperintermediate

Architecture of a Database System

Read on berkeley.edu

Summary

The paper surveys the internal architecture of modern relational DBMSs, describing the main components—client communications, process management, query processing, storage, and transaction management—and how they interact during a query lifecycle. It highlights design decisions such as admission control, buffer management, and concurrency control that are critical for scalability and reliability.

  • DBMS is built around five core modules: client communications manager, process manager, relational query processor, transactional storage manager, and shared utilities (catalog, memory manager).
  • Admission control in the process manager decides when a query can start based on resource availability, impacting throughput and latency.
  • The query processor compiles SQL into an operator tree that executes against storage, using a lock manager and log manager to enforce ACID properties.
  • Buffer management and access methods in the storage layer handle paging between disk and memory, crucial for performance.

DB engineers and architects need to understand these layers to design, tune, or extend a database system effectively.

6/10
storagearchitecturedatabasestransactionquery processing

Related reading

  1. The Internals of PostgreSQL

    The Internals of PostgreSQL is a continuously updated, chapter‑based documentation covering PostgreSQL’s subsystems—from process architecture to replication—targeted at developers and DBAs. It serves as a comprehensive reference but the article itself offers little new technical detail.

    Hall of Fameinterdb.jp7 min

  4. Database for AI Agents: 5 Evaluation Criteria

    Databricks outlines five criteria for a production‑ready database for AI agents—branch‑per‑agent isolation, serverless scale‑to‑zero, hybrid search, ACID guarantees, and a unified platform that eliminates ETL lag—illustrating each with features of its Lakebase offering and brief customer anecdotes.

    Databricksdatabricks.com10 min

  5. Article: Beyond Relevance: A Governance-First Architecture for Enterprise Personalization

    The article proposes a governance‑first architecture for enterprise personalization, where policy‑driven steps (memory, journey graph, AI routing, scoring, trust checks, outcome simulation) shape the recommendation before it is returned. A reference FastAPI implementation demonstrates the pattern with external YAML policies and optional LLM assistance.

    InfoQinfoq.com19 min

  6. The Tail at Scale

    The paper defines “tail‑tolerant” systems that keep interactive latency (≤100 ms) low even as services scale to thousands of servers, and surveys the main causes of latency outliers. It shows that techniques such as request replication, hedged requests, and resource isolation can dramatically shrink the latency tail with modest overhead, enabling higher utilization without over‑provisioning.

    Hall of Fameresearch.google1 minpaper