Ayder
HTTP-native durable event log written in C (curl as client)
Details
- External ID
- 46604862
- Source
- HN
- Company
- —
- Product
- Ayder
- Website domain
- github.com
- Launched
- Jan. 13, 2026
- Cohort
- —
- Upvotes
- 56
- Upvotes percentile
- 0.8208168642951251
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:25 p.m.
- Updated at
- Sept. 7, 2026, 9:25 p.m.
Description
Hi HN,I built Ayder — a single-binary, HTTP-native durable event log written in C. The wedge is simple: curl is the client (no JVM, no ZooKeeper, no thick client libs).There’s a 2-minute demo that starts with an unclean SIGKILL, then restarts and verifies offsets + data are still there.Numbers (3-node Raft, real network, sync-majority writes, 64B payload): ~50K msg/s sustained (wrk2 @ 50K req/s), client P99 ~3.46ms. Crash recovery after SIGKILL is ~40–50s with ~8M offsets.Repo link has the video, benchmarks, and quick start. I’m looking for a few early design partners (any event ingestion/streaming workload).
Enrichment
- Theme
- developer infrastructure and monitoring utilities
- Vertical
- Horizontal
- Function
- Data infrastructure
- Audience
- Developer
- AI stance
- Not AI
- Project type
- Commercial product
- Normalized one-liner
- http-native durable event log system
- Manually corrected
- False
Could you build this?
No Writing a durable, crash-consistent event log in C that survives SIGKILL with zero data corruption requires deep systems programming, storage engine design, and Linux kernel I/O expertise.
What it would actually take: The architecture requires low-level C with custom append-only segment storage, WAL, write barriers, fsync synchronization, memory mapping, and a custom epoll-based HTTP server. The core difficulty lies in crash recovery semantics, byte-level offset durability without external consensus engines, and extensive Jepsen-style fault injection testing.
Discussion
20 comments analyzed.
Competitors mentioned: Kafka, Feed API spec
Concerns raised: Performance comparable to quickly-written Python implementations, Poor quality software with bad performance (Kafka mentioned as established alternative), Overhead with indirection on cloud platforms like Digital Ocean, Scheduler overhead in userspace implementation, README feels like LLM-generated marketingspeak rather than authentic documentation
Feature requests: HTTP Range headers for offset-based pagination, Event log consumption as standardized protocol across brokers, Mature client libraries and external specification to avoid bikeshedding, Simpler Makefile using implicit C compilation rules
Competitors
Other products that read as similar to this one — 70 launches clear the similarity bar, closest 8 shown.
Attention rank: #13 of 71 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 73 days after the earliest competitor.
- Durable Streams · hn · 2025-12-09 · 10 upvotes · similarity 0.45
- An extensible pub/sub messaging server for edge applications · hn · 2026-01-28 · 44 upvotes · similarity 0.42
- I built an HTTP client that perfectly mimics Chrome 142 · hn · 2025-11-08 · 39 upvotes · similarity 0.41
- HTTP/3 and raw QUIC client/server APIs for Node.js · hn · 2026-06-08 · 16 upvotes · similarity 0.41
- Durable Endpoints · hn · 2026-02-18 · 8 upvotes · similarity 0.41
- UnisonDB · hn · 2025-11-01 · 17 upvotes · similarity 0.39
- RePlaya · hn · 2026-06-02 · 50 upvotes · similarity 0.38
- Verani · hn · 2025-12-12 · 6 upvotes · similarity 0.37
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a data infrastructure tool for Media & entertainment yet.