Fast and secure GitHub Actionson your own infrastructure.
Standard hosted runners are slow and expensive, larger runners cost even more.
Actuated runs on your own infrastructure with a predictable cost, low management and secure isolation with Firecracker.
Trusted by a growing number of companies
After one week actuated has been a huge improvement! Even our relatively fast Go build seems to be running 30%+ faster and being able to commit to several repos and only wait 10 minutes instead of an hour has been transformative.
Finally our self-hosted GitHub Actions Runners are on fire 🔥. Thanks to Alex for all the help onboarding our monorepo with NX, Docker, x86 and Arm on bare-metal.
Our worst case build/deploy went down from 1h to 20-30mins, for a repo with 200k lines-of-code.
Increase speed, security and efficiency.
Learn what friction actuated solves and how it compares to other solutions: Read the announcement
- Secure CI that feels like hosted.
Every job runs in an immutable microVM, just like hosted runners, but on your own servers.
That means you can run
sudo, Docker and Kubernetes directly, just like you do during development, there's no need for Docker-in-Docker (DIND), Kaniko or complex user-namespaces.What about management and scheduling? Provision a server with a minimal Operating System, then install the agent. We'll do the rest.
- Arm / M1 from Dev to Production.
Apple Silicon is a new favourite for developer workstations. Did you know that when your team run Docker, they're using an Arm VM?
With actuated, those same builds can be performed on Arm servers during CI, and even deployed to production on more efficient Ampere or AWS Graviton servers.
- Run directly within your datacenter.
If you work with large container images or datasets, then you'll benefit from having your CI run with direct local network access to your internal network.
This is not possible with hosted runners, and this is crucial for one of our customers who can regularly saturate a 10GbE link during GitHub Actions runs.
In contrast, VPNs are complex to manage, capped on speed, and incur costly egress charges.
- Debug live over SSH
We heard from many teams that they missed CircleCI's "debug this job" button, so we built it for you.
We realise you won't debug your jobs on a regular basis, but when you are stuck, and have to wait 15-20 minutes to get to the line you've changed in a job, debugging with a terminal can save you hours.
"How cool!!! you don't know how many hours I have lost on GitHub Actions without this." - Ivan Subotic, Swiss National Data and Service Center (DaSCH)
Lower your costs, and keep them there.
Actuated is billed by the maximum amount of concurrent jobs you can run, so no matter how many build minutes your team requires, the charge will not increase with usage.
In a recent interview, a lead SRE at UK-based scale-up told us that their bill had increased 5x over the past 6 months. They are now paying 5000 GBP / month and we worked out that we could make their builds faster and at least halve their costs.
- Lower management costs.
Whenever your team has to manage self-hosted runners, or explain non-standard tools like Kaniko to developers, you're losing money.
With actuated, you bring your own servers, install our agent, and we do the rest. Our VM image is built with automation and kept up to date, so you don't have to manage packages.
- Predictable billing.
Most small teams we interviewed were spending at least 1000-1500 USD / mo for slower hosted runners.
That cost would multiply if moved to GitHub's "bigger runners" and increas with usage. But with actuated the cost is flat-rate, no matter how many minutes you use.
Get insights into your organisation
When you have more than a few teammates and a dozen repositories, it's near impossible to get insights into patterns of usage.
Inspired by Google Analytics, Actuated contrasts usage for the current period vs the previous period - for the whole organisation, each repository and each developer.
Improve your builds with our tips & tricks
Learn how to improve the speed and efficiency your GitHub Actions with our examples and best practices.
These are the results you've been waiting for
“Our team loves actuated. It gave us a 3x speed increase on a Go repo that we build throughout the day. Alex and team are really responsive, and we're already happy customers of OpenFaaS and Inlets.”
“We just switched from Actions Runtime Controller to Actuated. It only took 30s create 5x isolated VMs, run the jobs and tear them down again inside our on-prem environment (no Docker socket mounting shenanigans)! Pretty impressive stuff.”
“From my time at Mirantis, I learned how slow and unreliable Docker In Docker can be. Compared to GitHub's 16 core runners, actuated is 2-3x faster for us.”
“This is great, perfect for jobs that take forever on normal GitHub runners. I love what Alex is doing here.”
“We needed to build public repos on Arm runners, but QEMU couldn't finish in 6 hours, so we had to turn the build off. Actuated now builds the same code in 4 minutes.”
“We needed to build Arm containers for our customers to deploy to Graviton instances on AWS EC2. Hosted runners with QEMU failed to finish within the 6 hour limit. With actuated it takes 20 minutes - and thanks to Firecracker, each build is safely isolated for our open source repositories.”
“One of our Julia builds was taking 5 hours of compute time to complete, and was costing us over 1500USD / mo. Alex's team got us a 3x improvement on speed and lowered our costs at the same time. Actuated is working great for us.”
Frequently asked questions
“It's cheaper - and less frustrating for your developers — to pay more for better hardware to keep your team on track.
The upfront cost for more CPU power pays off over time. And your developers will thank you.”
Read the article: The hidden costs of waiting on slow build times (GitHub.com)
Ready to go faster?
You can try out actuated on a monthly basis, with no commitment beyond that.
Make your builds faster today