Welcome to QSCS
On this page 2 sections
QSCS is a small, fast HTTP edge daemon that puts a smart cache in front of your origin web servers and replicates that cache across a cluster of nodes. It is designed to keep your site reachable through any failure mode, and to make distributed web applications dramatically simpler, faster, and more secure than a hand-rolled stack of load balancers, queues, and replication scripts.
Who this documentation is for
Anyone deploying QSCS in front of one or more websites. No specific cloud
provider, container runtime, or operating system is assumed, only that
you can run a small native binary on a Linux host.
What QSCS gives you
- Resilience. QSCS keeps a recent copy of every response it has seen and continues serving it during origin or peer failures (HEADLESS mode). Outages stop being visible to your users.
- Security. Every node speaks to its peers over a single encrypted port. Inter-node replication, control-plane chatter, and public traffic all flow over the same TLS-protected ingress, so the firewall surface of an entire global cluster is one port per node, not a list of replication and admin ports per service.
- Lower global latency. Thin clients hold a regionally local copy of the cache. Users hit the nearest edge over a short network path; only cache misses traverse the long link to the master.
- Less distributed-systems complexity. No quorum service, no shared database, no message broker. Joining a cluster is one command and the node receives its peers, role, and origins automatically.
- Observable. A loopback Prometheus-style metrics endpoint exposes everything you need to alert on.
Where to go next
- What is QSCS?, the one-page elevator pitch.
- Quick Start, install, subscribe, and verify in five minutes.
- Recommended Topologies, pick the deployment shape that matches your needs.
- Core Concepts, the small vocabulary you need to read the rest of the docs.