From cf3a50240f8d680bafaa9467e684774275969a98 Mon Sep 17 00:00:00 2001 From: Stephen Cefali Date: Tue, 18 Jun 2024 09:01:14 +0800 Subject: [PATCH] more updates to readme --- README.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 1f1b2b8f..c725e4b9 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,8 @@ -# Fly PgVector +# Fly Pgvector -Open-source vector similarity search for Postgres +Open-source vector similarity search for Postgres on Fly.io. This is meant to be deployed with GitHub Actions so it requires making a forked version of the repo in your own account. -Fly PgVector leverages the flexibility of Fly's [Postgres Flex](https://github.com/fly-apps/postgres-flex) as a base, integrating with its scalable and configurable environment to store and manage vector data efficiently. +Fly Pgvector leverages the flexibility of Fly's [Postgres Flex](https://github.com/fly-apps/postgres-flex) as a base, integrating with its scalable and configurable environment to store and manage vector data efficiently. Store your vectors with the rest of your data. Supports: @@ -17,20 +17,20 @@ Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recove ## Installation -### Fly Setup +### Fly Setup -To deploy pgvector on Fly, start by installing the Fly CLI and setting up your Fly account: +If you already configured the Fly CLI before, skip to step 3. 1. Download and install the Fly CLI from the [Fly official website](https://fly.io/docs/getting-started/installing-flyctl/). 2. Open your terminal and run `flyctl auth signup` to create a new Fly account or `flyctl auth login` to log into an existing account. -3. Once logged in, fork this repository and clone it to your local machine. Use the command `git clone git@github.com:scefali/fly-pgvector.git` followed by `cd fly-pgvector` to navigate into the project directory. +3. Fork this repository and clone it to your local machine. Use the command `git clone git@github.com:/fly-pgvector.git` followed by `cd fly-pgvector` to navigate into the project directory. 4. Create a new Fly Postgres app by running `fly pg`. This command sets up a new Postgres instance on Fly. Note at this stage this is still the base Fly Postgres image instead of the one with pgvector. 5. After setting up your Fly Postgres app, copy the output connection string provided. Store this string securely as you will need it to connect to your database. Consider using a password manager or another secure method to keep this information safe. 6. Enable GitHub Actions for your repository if not already enabled. Go to your repository on GitHub, click on 'Actions', then select 'set up a workflow yourself' if you haven't done so already. This will allow you to create a new workflow file directly or commit an existing one. 7. To authenticate your Fly CLI with the Fly platform, you need an API token. You can obtain this token by navigating to your Fly dashboard. Go to the 'Account Settings' and click on 'Access Tokens'. Create a new token if you don't have one already, and copy it. 8. To set the API token as an environment variable in GitHub Actions, navigate to your repository on GitHub. Go to 'Settings' > 'Secrets' > 'Actions'. Click on 'New repository secret'. Name the secret `FLY_API_TOKEN` and paste your copied token into the value field. This will securely store your token and make it available in your GitHub Actions workflows. 9. Configure your app by modifying the `fly.toml` file, which was generated during the app creation process, to suit your deployment needs. This file includes settings for regions, resources, and more. -10. Commit your code to master, push to GitHub, and the GitHub Action for deploying a new Fly version should run. +10. Commit your code to master, push to GitHub, and the GitHub Action for deploying a new version of Fly Pgvector should run. ### Making the Database accessible to outside world (optional)