Fission is building the Everywhere Computer
The Everywhere Computer (ϵ✵) lets you write complex code that runs everywhere — your local machine, proprietary hyper-clouds, and everything in between. It lets you compose small, powerful functions into complex workflow logic, and execute them reliably at scale.
When you choose the Everywhere Computer, you choose the ability to:
- Run your own self-managed network or rent from ours
- Write functions in your language of choice: Rust, JS, ++
- Compose functions and workflows written by others
- Extend the network’s capabilities with any UCAN-enabled service
Why choose the Everywhere Computer?
You already have many ways of running compute over a network. When is the Everything Computer a good fit? Choose the Everywhere Computer when…
…your project is threatened by:
- Low Connectivity: Because the Everywhere Computer runs local-first, you can continue to process compute at the edge with the resources you have at hand. As peers rendevouz or connectivity improves, that’ll just become smoother and faster.
- High Data Gravity: A lot of data in one spot becomes cumbersome to move. We call that “data gravity”. The Everywhere Computer lets compute move to your data, instead of the other way around.
- Narrow Bandwidth Constraints: Sometimes you simply can’t run all the analysis you need on your data at the edge. The Everything Computer can be used to pre-process and select only the most important bits to send over the wire when bandwidth is scarce.
- Low Network Throughput: Blockchains are fantastic example of this: for certain things that need global, timely consensus they work great. But not everything does. For everything else, the Everywhere Computer can deliver the verifiability you needs at a fraction of the time & cost.
- Regulatory Compliance: It’s critical for your project to comply with regulations, but that often comes at an operational cost. It takes time and money to do achieve compliance. Whatever your compliance needs — uptime guarantees, auditability, etc — the Everywhere Computer is less resource intensive than alternatives.
- Platform Risk: In an environment where vendor lock-in and platform dependencies are common, the Everywhere Computer offers a decentralized and open alternative. This reduces your reliance on any single provider and mitigates risks associated with platform-specific changes, outages, or discontinuation of services.
…your project needs:
- Multi-cloud Deployments: Whether for the regulatory compliance mentioned above, or because of legacy code or newly acquired codebases, multi-cloud deployments are a reality for many projects. Portability of the Everywhere Computer enables your company to make better use of resources and deploy a single codebase to multiple targets.
- Mission Critical Reliability: Control systems in mission critical environments will benefit from the Everywhere Computer’s durability — replaying execution from the point of failure, and running the execution in parallel. Your software will be resilient to outages and protected against data loss.
- Reproducible Computation: The Everywhere Computer ensures that computations can be consistently replicated across different environments. This is crucial for scientific research, data analysis, and any application where consistency and reliability of computational results are paramount.
- Community Maintenance: In a project where diverse community members collectively guide and care for the system, the Everywhere Computer’s portability and durability shine. It can enable all members of the community to scale their participation up and down as their resources, interest and abilities permit.
- Credible Exit: With the Everywhere Computer, you’re not locked into a single vendor or platform, ensuring a credible exit strategy. This flexibility allows you to adapt to changing technological landscapes without the risk of being tethered to a specific vendor or technology.
In short: the Everywhere Computer can provide value any time that the portability or durability of the code you run are primary concerns.
Who participates in the Everywhere Computer?
There are many ways to get involved.
As an Operator
Because they run everywhere, Homestar nodes are ready to be operated at a wide array of scales. We encourage you to do so! At any scale, operating your nodes privately or connecting into the open Everywhere Computer network is your choice.
- Home-scale: Run a Homestar node on your local machine or personal cloud account as we boot up the network. No matter how you participate in the Everywhere Computer, you’ll start as a home-scale operator. And it’s easy to do.
- Area-scale: eg. factories, warehouses, neighbourhoods, cities
- World-scale: eg. hyperclouds, telco infrastructure, logistics networks, global commons
As a Function Author
Try your hand at writing functions and workflow templates for the Everywhere Computer. They compile to Wasm (WebAssembly) and can be run by anyone in the network.
- Front-End Devs: Get backend capabilities that run local-first, on your users’ devices, and run at scale on the Everywhere Computer when needed.
- Rust(aceans): Rust is the native language of our Homestar nodes, and is ready for you to author in today.
- BYOL (Bring Your Own Language): For languages that support Wasm & the Wasm component model, you can write and deploy to the Everywhere Computer today. And, for those that don’t, the path is open for you to develop a language integration on your own by supporting Wasm & the Wasm component model.
As a Service Provider
Beyond authoring functions, and operating nodes, there are a variety of services participants in the network will need.
- Consulting: Auditing, workflow optimization and custom functions are just a few examples.
- Tooling: Curated registries, observability platforms and IDE integrations.
- Cloud Services (aka IaaS, BaaS, PaaS and SaaS): The Everywhere Computer is compatible with any UCAN-enabled service endpoint.
Built on open protocols
IPVM, the InterPlanetary Virtual Machine, aims to be the easiest, fastest, most secure, and open way to run WebAssembly functions anywhere.
Using content-addressed data, public key infrastructure (PKI), and capabilities, IPVM liberates computation from its dependence on pre-negotiated services and paves the way for interoperability.
IPVM is designed to run an open, community standards base of portable compute functions, instead of being locked into any one cloud provider.
Homestar is Fission's reference implementation of the IPVM standards, written in Rust. The managed network we're going to run is what we call the Everywhere Computer.
Join the early access beta
If you'd like to participate in the Everywhere Computer beta, please complete this form.Beta intake form 🌐 💻
While you're await for your beta access, here's some places to learn more:
- Watch our presentations on IPVM, Homestar and our vision for the Everywhere Computer
- PL Summit pitch: an update on IPVM and launch of Everywhere Computer
- IPFS Connect: talking a bit more detail about IPVM as protocol and Everywhere Computer as platform
- The Edge of Tomorrow: CTO Brooklyn Zelenka's Lightning Talk covering the vision for IPVM and what has become the Everywhere Computer at ETHCC Blueyard, 2022
- Read through the docs
- Dive into the Homestar code on Github
- Join the IPVM working group for information on getting involved in the standards process.
- Hop into the Fission Discord and introduce yourself 🤝. Once you're a part of the Beta, you'll be added to a private beta forum.