Free. As In Beer.
If you never need to queue‐up lots of messages because your applications are super‐efficient, you’re not going to need more. So, we’ll be free. If you need more, simply pay monthly for the resources you need. We don’t charge for bandwidth – just connections and message storage. You can downgrade at any time.
Unlimited Monthly Messages
Send. Store. Receive. Any size. Any number: 1 or 1,000,000,000. Just pay for the storage you need. Consume messages quickly? Then storage is free.
When you Sign Up, you’ll be allocated a free quota of disk space, RAM and connections. You allocate this to store your messages (or keep them in memory) and manage environments, users, durable subscriptions and the like. If you promptly consume your messages, your queues will never have a lot of messages stored in them, and you’ll never exceed your free quota!
Open Messaging: AMQP 1‧0
What’s Cloud Computing’s biggest weakness? Not security, but lack on interoperability. Using open protocols, like AMQP, means you can switch to us from in‐house products – and switch away from us without hidden penalty. That means you can realise Cloud Computing’s biggest win: instant commoditisation.
Our technology includes several trade secrets. Like how to zero‐copy multiplexed, partial message data destined to multiple queues. Or use probabilistic non‐blocking priority queue access algorithms. How to make each connection sticky to a CPU. Or manage 100,000s of timed events (hint to competitors: try cutting a pack of cards using a log).
PostgreSQL Admin Interface
Need to automate or integrate? Our administration API is SQL compatible. Create entire environments or tweak a queue from the command line. The choice is yours.
We want you to leverage the skills you’ve got. Using PostgreSQL for our administration API means you can integrate into your dashboards and processes using tools you already love – or you can just batch jobs using bash from the command shell. Far easier than learning a whole new bespoke REST API.
Store & Forward
How reliable would you like your messaging? We support from At‐Most‐Once through At‐Least‐Once to Exactly‐Once. Worried that messages should be persisted? No problem (persisted message quotas are cheaper, too). Even if we upgrade, we don’t lose messages in memory, either. Unreliable network? We support partial re‐delivery of fragments, so you can pick up where you left off. Perfect for big messages over bad connections. Send lots of messages and want to optimise acknowledgments – fine. Want to keep the state of a connection for next time? Certainly, just use part of your quota.
Publish & Subscribe
For us, a topic is a queue and a subscription. Filter by what makes sense to you. Queues can last for ever, or just for the lifetime of a connection. Need a queue to self‐destruct when it’s empty? No problem. What about when there are no subscribers? Easy.
IT Audit Ready
Cloud Computing should be compatible with the best IT audit practices. We should know – we’ve been on the other side. We record everything in excruciating detail, so when you need to complete a PCI DSS audit, we’ll be the least of your concerns. Multi‐tenancy is essential for true Cloud Computing. It’s impossible to do securely unless it’s done from scratch. So we did it using our own custom memory pool allocators atop malloc.
Sandbox your set‑up
Create and isolate resources by system or environment. Need development to have the same roles and queues but different access? Done.
You get your unique company space, in which you can have as many systems and environments (development, production, etc) as you like. Your messages are stored separately from not only everyone else but from each environment configuration. To ease release management, each environment can inherit its own quota of resources but have have the same logical roles and permissions – with different passwords, naturally. Administrators can manage multiple environments or systems.
Secure by Default
Whole‑disk encryption. 2048‑bit TLS 1.2 connections. SASL authentication integration. Internal VPNs. Fine‑grained permissions.
What happens if one customer gets hacked? It’s not just your reputation lost – it’s ours too. So we built security in from the OS upwards. Firewalls are integrated into every server. Network traffic is always encrypted, even in our data centre, in every rack. SASL mechanisms for deep integration into your security model are possible – just ask. We use external white hat consultants to pen‐test our service and identify vulnerabilities. Oh, and you can’t choose your users’ passwords. We do, so they have enough entropy.