Skip to content

Commit bf61a84

Browse files
committed
update docs
1 parent 75db5fe commit bf61a84

11 files changed

Lines changed: 106 additions & 5114 deletions

.gitignore

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -10,3 +10,4 @@ website/.next/
1010
website/.source/
1111
.omx/
1212
.local
13+
.pista

README.md

Lines changed: 19 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ Tela is the latin word for "Fabric", which refers to the interconnected network
99

1010
## Latest Updates
1111

12-
* **[2026/02]** 💡 **How SwissAI Leverages OpenTela**: We wrote a case study on how SwissAI uses OpenTela to orchestrate their distributed GPU nodes for scalable model serving. [Read more](docs/posts/swissai.md).
12+
* **[2026/02]** 💡 **How SwissAI Leverages OpenTela**: We wrote a case study on how SwissAI uses OpenTela to orchestrate their distributed GPU nodes for scalable model serving. [Read more]((docs/posts/swissai.md).
1313

1414
## Features
1515

@@ -25,10 +25,24 @@ Tela is the latin word for "Fabric", which refers to the interconnected network
2525

2626
## Documentation
2727

28-
- [Installation](docs/tutorial/installation.md)
29-
- [Spin Up a network](docs/tutorial/spinup.md)
30-
- [How requests are routed](docs/tutorial/routing.md)
31-
- [Glossary](docs/tutorial/glossary.md)
28+
### Getting Started
29+
- [Installation](docs/tutorial/installation) — Download and install OpenTela
30+
- [Spin Up LLM Serving](docs/tutorial/spinup) — Set up multi-LLM serving cluster
31+
- [Request Routing](docs/tutorial/routing) — Understand how requests are routed
32+
- [Wallet & Ownership](docs/tutorial/owner) — Manage Solana wallets and node identity
33+
- [Solana Settlement](docs/tutorial/settlement) — Configure automated usage billing
34+
- [Docker Serving](docs/tutorial/docker-serving) — Use Docker containers for LLM serving
35+
- [Glossary](docs/tutorial/glossary) — Key terms and concepts
36+
37+
### Advanced Topics
38+
- [CRDT Internals](docs/advanced/crdt-internals) — How CRDT synchronization works
39+
- [CRDT Tombstones](docs/advanced/crdt-tombstones) — Node departure handling
40+
- [Security Hardening](docs/advanced/security) — Build attestation, trust, and access control
41+
- [Performance Benchmark](docs/advanced/performance-optimization) — Proxy latency measurements
42+
- [Large-Scale Simulation](docs/advanced/benchmark) — Run 100+ node simulations
43+
44+
### Extensions
45+
- [Fleet Manager](docs/extensions/fleet-manager) — Deploy to SLURM clusters with otela-fleet
3246

3347
## Contributing
3448

docs/README.md

Lines changed: 79 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,79 @@
1+
# OpenTela
2+
3+
A decentralized distributed computing platform that orchestrates GPU resources across a peer-to-peer network for LLM serving.
4+
5+
## Overview
6+
7+
OpenTela connects GPU resources into a unified pool using:
8+
- **libp2p networking** for peer-to-peer communication
9+
- **CRDT-based state management** for distributed consensus
10+
- **Identity group routing** for intelligent request distribution
11+
- **Solana settlement** for automated usage billing
12+
13+
Primary use case: Distributed GPU node orchestration for LLM serving, powering projects like the [SwissAI Initiative](https://serving.swissai.cscs.ch/).
14+
15+
## Quick Start
16+
17+
### Installation
18+
19+
```bash
20+
# x86_64
21+
wget https://github.com/eth-easl/OpenTela/releases/latest/download/otela-amd64 -O otela && chmod +x otela
22+
23+
# arm64
24+
wget https://github.com/eth-easl/OpenTela/releases/latest/download/otela-arm64 -O otela && chmod +x otela
25+
```
26+
27+
### Spin Up a Cluster
28+
29+
**Head Node:**
30+
```bash
31+
./otela start --mode standalone --public-addr {YOUR_IP} --seed 0
32+
```
33+
34+
**Worker Node:**
35+
```bash
36+
./otela start \
37+
--bootstrap.addr /ip4/{HEAD_IP}/tcp/43905/p2p/{HEAD_PEER_ID} \
38+
--subprocess "vllm serve Qwen/Qwen3-8B --port 8080" \
39+
--service.name llm \
40+
--service.port 8080
41+
```
42+
43+
### Send Requests
44+
45+
```python
46+
import openai
47+
client = openai.OpenAI(
48+
base_url="http://{HEAD_IP}:8092/v1/service/llm/v1",
49+
api_key="test-token"
50+
)
51+
response = client.chat.completions.create(
52+
model="Qwen/Qwen3-8B",
53+
messages=[{"role": "user", "content": "Hello!"}]
54+
)
55+
```
56+
57+
## Documentation
58+
59+
Full documentation is available at the [OpenTela Docs](https://docs.opentela.ai) or browse the [`content/docs/`](content/docs/) directory:
60+
61+
- **Tutorial** — Installation, spinup, routing, wallets, settlement
62+
- **Advanced** — CRDT internals, performance, security
63+
- **Blog** — Real-world deployments (SwissAI)
64+
- **Proposals** — Design documents
65+
66+
## Development
67+
68+
This repository contains the documentation site built with Next.js and Fumadocs.
69+
70+
```bash
71+
npm install
72+
npm run dev
73+
```
74+
75+
The OpenTela binary source code is available at [eth-easl/OpenTela](https://github.com/eth-easl/OpenTela).
76+
77+
## License
78+
79+
MIT

docs/content/docs/extensions/fleet-manager.md

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,8 @@
1+
---
2+
title: Fleet Manager
3+
description: Launch, inspect, and reconcile serving workloads across SLURM-backed clusters with otela-fleet.
4+
---
5+
16
# Fleet Manager
27

38
The OpenTela fleet manager helps you launch, inspect, and reconcile serving

docs/content/docs/tutorial/installation.mdx

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -32,8 +32,8 @@ Available Commands:
3232
help Help about any command
3333
init Initialize OpenTela: create config directory and a default wallet
3434
start Start listening for incoming connections
35-
update Update the Open Compute Binary
36-
version Print the version of otela
35+
update Update the OpenTela Binary
36+
version Print the version of OpenTela
3737
wallet Wallet management commands
3838

3939
Flags:

0 commit comments

Comments
 (0)