SShortSingh.
Back to feed

Developer Tests PgCache Proxy on 16-Million-Row Postgres DB, Finds Real Speed Gains

0
·1 views

A developer built a synthetic 16-million-row e-commerce database to benchmark PgCache, a Postgres proxy that caches query results and invalidates them via the database's replication stream rather than using timers or manual cache-busting logic. The setup required only enabling logical replication on Postgres and adjusting a shared-memory setting, after which PgCache automatically created its own publication and replication slot. Unlike materialized views or Redis-based caching, PgCache sits transparently in front of Postgres using the wire protocol, meaning applications need only change the port they connect to. Benchmarks ran four aggregate queries 150 times each at concurrency 10, comparing a properly indexed Postgres instance directly against the same instance fronted by PgCache. The project, tested on PgCache version 0.6.2, is fully reproducible via a public GitHub repository using Docker Compose.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

How Apache Iceberg's Puffin Format Solves Query Optimization Gaps in Large Tables

Apache Iceberg's Puffin is a lightweight container file format designed to store statistics that do not fit inside Iceberg's standard manifest files. Manifests efficiently hold per-file scalar metrics like row counts and value bounds, but cannot accommodate larger structures such as distinct-value sketches or row-level delete bitmaps. Without distinct-value estimates, query optimizers are forced to guess join strategies, potentially causing unnecessary shuffling of terabytes of data. Puffin addresses this by housing two blob types: Theta sketches for estimating distinct column values, and deletion vectors for tracking row-level deletes introduced in Iceberg format version 3. The format uses a simple structure — a magic number, opaque data blobs, and a JSON footer — allowing query engines to access costly statistics without bloating the manifests read on every query.

0
ProgrammingDEV Community ·

OpenAI Launches Prism, a Free AI-Powered LaTeX Workspace for Researchers

OpenAI has introduced Prism, a free cloud-based writing workspace built on its GPT-5.2 model, designed to help scientists draft, revise, and prepare research papers for publication. The platform combines a LaTeX editor with document-aware AI assistance, keeping equations, citations, figures, and references within a single working environment. Prism evolved from Crixet, a cloud-based LaTeX platform that OpenAI previously acquired and redeveloped. The workspace is currently available at no cost to personal ChatGPT account holders across Free, Go, Plus, and Pro tiers, with no stated seat limits for personal use. OpenAI has indicated that paid features and expanded access for Business, Enterprise, and Education users are planned, though no pricing details or firm rollout dates have been announced.

0
ProgrammingDEV Community ·

Why Measuring Pull Requests May Be Hurting Your Software Team's Performance

A software engineering advocate argues that measuring pull request metrics—such as duration and throughput—is counterproductive because PRs inherently slow down code integration and delivery. Citing a DX study of over 500 organizations, the author notes that the longer PRs remain open, the less code developers ship, particularly in high-performing teams. Rather than optimizing a practice that introduces delays, the author urges engineering leaders to adopt DORA metrics—deployment frequency, lead time for changes, change failure rate, and time to restore service—as more meaningful indicators of delivery performance. The piece contends that continuous integration with trunk-based development is a superior alternative, supported by over a decade of DORA research showing that speed and stability improve together. Using a marathon analogy, the author warns that measuring and rewarding PR efficiency is akin to timing tea service during a race—optimizing a stop that should not exist in the first place.