github-personal/firezone
1
0
Fork 0
mirror of https://github.com/outbackdingo/firezone.git synced 2026-01-27 10:18:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(docs): Add link to Terraform examples (#4508)

Browse Source 
Fixes #2620
This commit is contained in:
Jamil
2024-04-04 10:00:15 -07:00
committed by GitHub
parent a58dd99e9f
commit 9557dbc4ae
1 changed files with 30 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

58
website/src/app/kb/deploy/gateways/readme.mdx
View File

@@ -59,34 +59,30 @@ deployment.
correctly.
</Alert>
### Sizing recommendations
Gateways, like the rest of Firezone's data plane stack, are written in Rust and
are thus Resource efficient by nature.
A single Client connection to a single Gateway can typically reach speeds of **1
Gbps** or more. This scales linearly with each Client connection up to the
number of CPU cores available to the Gateway.
In our tests, we've found that even a single `f1.micro` instance on Google Cloud
Platform, equipped with a single shared vCPU core, can effortlessly handle up to
1 Gbps of tunneled traffic. This performance is attributed to the efficiency of
the WireGuard® protocol and the Rust programming language.
Use the table below as a rough guide for sizing your Gateway deployments. These
are general recommendations and may not be accurate for your specific use case.
If you're unsure, start with a smaller Gateway and scale up as needed.
| Gateway size | Users served | CPU cores | Memory | Network link |
| ------------ | -------------- | --------- | ------ | ------------ |
| Micro | 10 - 100 | 1 | 512 MB | 1 Gbps |
| Small | 100 - 250 | 2 | 1 GB | 1 Gbps |
| Medium | 250 - 1,000 | 4 | 4 GB | 10 Gbps |
| Large | 1,000 - 10,000 | 16 | 16 GB | 25 Gbps |
To go beyond the table above, you can deploy additional Gateways and use
Firezone's [automatic load balancing](#load-balancing) to distribute Client
connections across them.
{/* ### Sizing recommendations */} {/* */}
{/* Gateways, like the rest of Firezone's data plane stack, are written in Rust and */}
{/* are thus Resource efficient by nature. */} {/* */}
{/* A single Client connection to a single Gateway can typically reach speeds of **1 */}
{/* Gbps** or more. This scales linearly with each Client connection up to the */}
{/* number of CPU cores available to the Gateway. */} {/* */}
{/* In our tests, we've found that even a single `f1.micro` instance on Google Cloud */}
{/* Platform, equipped with a single shared vCPU core, can effortlessly handle up to */}
{/* 1 Gbps of tunneled traffic. This performance is attributed to the efficiency of */}
{/* the WireGuard® protocol and the Rust programming language. */} {/* */}
{/* Use the table below as a rough guide for sizing your Gateway deployments. These */}
{/* are general recommendations and may not be accurate for your specific use case. */}
{/* If you're unsure, start with a smaller Gateway and scale up as needed. */}
{/* */}
{/* | Gateway size | Users served | CPU cores | Memory | Network link | */}
{/* | ------------ | -------------- | --------- | ------ | ------------ | */}
{/* | Micro | 10 - 100 | 1 | 512 MB | 1 Gbps | */}
{/* | Small | 100 - 250 | 2 | 1 GB | 1 Gbps | */}
{/* | Medium | 250 - 1,000 | 4 | 4 GB | 10 Gbps | */}
{/* | Large | 1,000 - 10,000 | 16 | 16 GB | 25 Gbps | */}
{/* */}
{/* To go beyond the table above, you can deploy additional Gateways and use */}
{/* Firezone's [automatic load balancing](#load-balancing) to distribute Client */}
{/* connections across them. */}
## Deploy a single Gateway
@@ -121,6 +117,12 @@ balancing.
portal for audit trail and logging purposes.
</Alert>
### Deploy using Terraform
See our
[Terraform examples](https://github.com/firezone/firezone/tree/main/terraform/examples)
for deploying and scaling Gateways on Google Cloud Platform.
## High availability
Firezone was designed from the ground up to support high availability