Skip to content
NewCache tags and signed URLs for every website

Your app, your files and your traffic, served from the edge

RivoCloud puts a global cache in front of your site, stores your files in S3-compatible buckets and shows you every request in real time. One console, one API, no glue code.

See how it works
  • Purges reach every edge in seconds
  • S3-compatible storage built in
  • Automatic TLS and signed URLs

Edge network · live traffic

RivoCloud edge networkA world map with fifteen edge locations. Visitors connect to their nearest edge, which answers from cache and only contacts the origin server on a cache miss.

Trusted by teams who ship every day

  • Acme Media
  • Northwind Shop
  • Lumen Labs
  • Pinecrest
  • Orbitly
  • Fieldnote
Edge locations on five continents
15
Requests served every day
2.4B
Median cache hit ratio
94%
Edge uptime over the last 12 months
99.99%

Customer names and figures on this page are illustrative.

Products

Three products, one console

Websites, storage and analytics share the same edge network, the same team and the same API. Pick a product to see it in action.

Cache your site at the edge, on your terms

Add a website by pointing RivoCloud at your origin. Every edge answers from memory first, then a shared Redis layer, and only goes back to your server when it has to.

  • Per-path rules: cache, bypass or no-store, with separate edge and browser TTLs
  • Choose whether query strings are part of the cache key
  • Purge by path pattern or cache tag; every edge drops it within seconds
  • Three cache layers keep origin traffic to a trickle

shop.northwind.com · Cache rules

4 rules
PathActionEdge TTL
/static/*Cache30 days
/products/*Cache10 min
/api/*Bypass-
/account/*No store-
Where requests were answeredlast 24 hours
  • Edge memory 71%
  • Regional Redis 23%
  • Origin 6%
/products/summer-*
Purge

Purged on all 15 edges in 1.4 s · 2,318 objects

How it works

From origin to edge in four steps

Most teams are serving traffic through RivoCloud within an hour. Select a step to see what happens behind the scenes.

Step 1 of 4

Active path
VisitorsDNSCNAME to rivocloud.devConsole and API/api/v1/zonesEdge location, nearest to the visitorEdge proxyTLS, signed URLs, rate limitsL1 memory~1 msL2 Redis~3 msMemory first, then Redis, then originYour originorigin.example.comMessage busconfig and purges, NATS+12Every other edgeAnalyticsClickHouse

The console saves the website and publishes its config to the edges, which will fetch from your origin on a miss.

Use cases

How teams use RivoCloud

The same three building blocks, put together in different ways. Here is how four kinds of teams set it up.

  • For example, Northwind Shop

    Online stores

    Keep product pages fast during a sale without caching anyone’s basket.

    How to set it up

    1. 1Add your store as a website with your shop server as the origin.
    2. 2Cache /static/* for 30 days and /products/* for 10 minutes.
    3. 3Set /cart/* and /account/* to no-store.
    4. 4Purge /products/* from your admin when prices change.
  • For example, Acme Media

    Media publishers

    Serve images and video from the edge and update a story the moment it changes.

    How to set it up

    1. 1Upload images and video to a storage bucket.
    2. 2Serve the bucket through your media domain so it is cached at the edge.
    3. 3Tag each article’s responses with a cache tag.
    4. 4Purge the tag when an editor publishes a correction.
  • For example, Lumen Labs

    SaaS apps

    Let customers upload files and share them privately, without building a file service.

    How to set it up

    1. 1Create a private bucket for customer uploads.
    2. 2Request a presigned URL from the API and upload from the browser.
    3. 3Turn on signed URLs for the website that serves the files.
    4. 4Hand customers links that expire after 15 minutes.
  • For example, Orbitly

    Developers and APIs

    Automate everything from CI: deploy, purge and check the numbers.

    How to set it up

    1. 1Create an API key scoped to zone:read and cache:invalidate.
    2. 2Store it as a secret in your CI pipeline.
    3. 3Purge the changed paths after every deploy.
    4. 4Read cache hit ratio from the analytics API to catch regressions.

Developers

Everything in the console is an API call

A REST API at /api/v1 covers websites, cache rules, purges, buckets, objects and analytics. A JavaScript SDK and a CLI are in preview.

  • API keys with scopes

    Give each script only what it needs. A deploy job that purges the cache never has to be able to delete a bucket.

  • An audit log for every change

    Changes made with a key are recorded with the key’s name, next to the ones your team makes in the console.

  • Predictable limits

    Sensitive endpoints such as purges are rate limited, and return a clear 429 when you need to slow down.

Available scopes include

  • zone:read
  • zone:write
  • cache:invalidate
  • storage:read
  • storage:write
  • analytics:read
curl https://api.rivocloud.dev/api/v1/zones \  -H "X-API-Key: $RIVO_API_KEY" \  -H "Content-Type: application/json" \  -d '{    "name": "northwind-shop",    "originUrl": "https://origin.northwind.com",    "cacheTtl": 3600  }'

Authenticate with an X-API-Key header. Keys start with rc_live_ or rc_test_.

Analytics

Know how fast you are, everywhere

Every request is logged at the edge and queryable within seconds. Switch between metrics and ranges below to explore sample traffic from a busy online store.

Total
12.89M
Peak per point
907K
Average per point
537K
0250K500K750K1.0M00:0011:0023:00

Response time at the edge

Time to first byte, last 24 hours

  • p5018 ms
  • p9541 ms
  • p9987 ms

Top paths

Share of all requests

  • /static/app.js21.4%
  • /products/*17.9%
  • /images/hero.webp9.3%
  • /api/cart4.1%

Sample data for illustration.

Customers

Teams ship faster with RivoCloud

  • Our product pages used to fall over every time we ran a sale. Now 95% of requests never reach our servers, and when prices change we purge one path from the admin and it is live everywhere before the page reloads.
    Jane DoeHead of engineering, Northwind Shop
  • We moved 30 TB of images and video into RivoCloud buckets in a weekend. Cache tags mean an editor’s correction is visible in seconds, not after a TTL runs out.
    John SmithPlatform lead, Acme Media
  • Presigned uploads and signed links replaced a file service we had been maintaining for three years. Our CI key can purge the cache and nothing else, which our security team loves.
    Alex JohnsonCTO, Lumen Labs

FAQ

Questions, answered

Want to see it for yourself? Try the demo console.

  • A pull zone is how RivoCloud represents one of your websites. You give it your origin URL, and our edges fetch ("pull") content from that origin the first time it is requested, then serve it from cache until it expires or you purge it. In the console we simply call it a website.

The network, by the numbers

Every website, bucket and cache rule adds to the same global network. Open the demo console to see the same numbers for a single account.

  • A seeded console with websites, cache rules, buckets and analytics
  • Four roles to try: owner, admin, developer and viewer
  • Runs in your browser, with no card and nothing to install

Network metrics

Last 30 days, across every edge location.

Live
Requests served
72.4B+14%
Bandwidth saved by cache
1.3 PB+9%
Median time to first byte
18 ms-3 ms
Websites on the network
3,200++210
Files stored in buckets
41 TB+6%
Global cache purge
<150 ms-20 ms

Figures are illustrative demo data.

Open the demo console