April 2026 · San Francisco
Three Services. One Database.
Gold Lapel Announces
Full Redis and Elasticsearch Parity.
“Teams are running three services — a cache, a search cluster, and a database —
to do what one PostgreSQL instance handles admirably.
Gold Lapel simply arranges the table so the capability is easy to reach.”
— The Butler of Gold Lapel
Self-optimizing PostgreSQL proxy delivers caching, queues, pub/sub, full-text search, vector similarity, and aggregations — with enterprise SSO, compliance exports, and distributed mesh networking. 34 methods across 7 languages. Enterprise-ready, solo-dev friendly.
Gold Lapel, the self-optimizing PostgreSQL proxy, today announces complete feature parity with both Redis and Elasticsearch. Caching, queues, pub/sub, full-text search, vector similarity, aggregations, and more — delivered directly from PostgreSQL through 34 methods across seven languages. No additional services. No data synchronization. No separate infrastructure to manage.
This is a considerable expansion from Gold Lapel's initial launch in March, which introduced automatic query optimization and a three-tier cache. The proxy now replaces three services with one database — and does so, if I may say, rather thoroughly.
The Redis Repertoire
Twenty-one methods across seven language wrappers replace the capabilities teams have long relied on Redis to provide. Each method builds upon what Redis pioneered with the addition of ACID guarantees, persistence by default, and the full power of SQL.
Caching — Three tiers: L1 in-process at 2µs, L2 proxy at 0.5ms, L3 materialized views at 0.7ms. All automatically invalidated on writes.
Pub/Sub — LISTEN/NOTIFY with wrapper-level subscribe and publish methods.
Queues — SKIP LOCKED-based job queues with enqueue and dequeue.
Sessions — Database-backed session storage with automatic expiry.
Rate Limiting — Sliding window rate limiters.
Data Structures — Sorted sets, hashes, geo operations, counters.
Lua Scripting — PL/Lua stored procedures via script().
The Elasticsearch Arrangements
Thirteen search methods across seven language wrappers replace the capabilities teams have long relied on Elasticsearch to provide. Each method builds upon what Elasticsearch established with the addition of row-level security, SQL expressiveness, and transactional consistency.
Full-Text Search — search() with tsvector/tsquery, 30+ language configurations, ts_rank scoring, highlighted snippets.
Fuzzy Matching — search_fuzzy() with pg_trgm trigram similarity.
Phonetic Search — search_phonetic() with Soundex and Double Metaphone.
Vector Similarity — similar() with pgvector, cosine distance, HNSW indexes.
Autocomplete — suggest() with prefix matching and similarity ranking.
Aggregations — facets() and aggregate() with SQL's window functions, JOINs, and CTEs.
Reverse Search — percolate_add(), percolate(), percolate_delete() for alerts, classification, and routing.
Custom Analyzers — create_search_config() for custom tokenization pipelines.
Relevance Tuning — analyze() and explain_score() for debugging search results.
Enterprise Features
SSO/SAML Authentication — OIDC and SAML 2.0 dashboard authentication, integrating with Okta, Azure AD, Google Workspace, and any compliant identity provider.
Compliance Audit Exports — JSON Lines and CSV exports of all optimization decisions, filterable by date and type, for SOC 2 and HIPAA compliance workflows.
Mesh Networking — Proxies auto-discover each other and propagate cache invalidation across instances in real time via encrypted P2P connections.
Citus Support — Auto-detects distributed PostgreSQL, propagates indexes to all shards automatically.
Serverless Detection — Auto-tunes for Neon, Supabase, Aurora Serverless, and other serverless PostgreSQL providers.
Cache Invalidation, Handled
Gold Lapel's NOTIFY-based write detection is enabled by default. Writes from any source — migrations, direct queries, companion services — are detected and trigger automatic cache invalidation across all three tiers. No TTLs. No manual key management. The matter, as ever, is attended to.
By the Numbers
34 methods — 21 comparable to Redis, 13 comparable to Elasticsearch.
7 languages, 4 frameworks, 3 ORMs — Python, Node.js, Ruby, Go, Java, PHP, .NET; Django, Rails, Spring Boot, Laravel; Prisma, SQLAlchemy, Drizzle.
The Books
Gold Lapel has published You Don't Need Redis and is preparing You Don't Need Elasticsearch — the first two books in a series that makes the case, thoroughly and without reservation, that most applications need only one database. Authored by The Butler of Gold Lapel.
Pricing
Usage-based, no tiers. $49/month per database, $49/month per mesh node, $5 per additional million queries (first million included, compliments of the butler). 14-day free trial. Enterprise pricing available for organizations requiring SSO, compliance exports, and dedicated support.
About Gold Lapel
Gold Lapel builds tools that help PostgreSQL serve as the only database service most applications need. The Butler of Gold Lapel serves all the needs of the household. The Groundskeeper of Gold Lapel is Stephen Gibson. The estate is based in San Francisco.
###
Media Contact
The Butler of Gold Lapel
thebutler@goldlapel.com
goldlapel.com