──No Cold Starts
──Scale pods to zero without unscheduling, and wake or migrate them instantly.
──Architect eliminates the tax on idle capacity with no code changes or complicated analytics suites.
> Tired of Performance Insurance?
Most clusters are deployed with extra capacity so that services are available when needed or to accommodate a traffic spike. This over-provisioning is expensive and wasteful.
Architect gives you zero-overhead pod snapshots and keeps pods scheduled and "running" in Kubernetes upon sleep. On wake, it eliminates delays associated with PVC mounts, load balancer registration, Kubernetes scheduling and software startup times associated with enterprise and stateful workloads so applications and services are available immediately.
Architect eliminates the need for over-provisioning and allows you to:
Hibernate any pod in place and wake it without cold starts in <50ms.
Use spot to deploy pods without service disruption upon pre-emption.
Move any pod anywhere with state integrity and without service interruption.
# How Architect works
Pod hibernation might seem simple, but waking one instantly is complex enough to be approaching magic. Waking a pod anywhere requires you to maintain complete state, reattach dependencies and maintain network connectivity. Fast and transparent.
We refer to Architect as a Kubernetes Availability Automation because it does all of this with a quick and simple install:
With a simple change to your deployment and ZERO code modification, you can employ Architect on any cluster for any pod.
The state of every pod in your cluster will be efficiently captured and ready to be “re”deployed or saved for hibernation.
Architect will automatically sleep any pod. Sound dangerous? We give you the ability to tune how Architect behaves based on your needs.
Upon wake, Architect instantly restores the pod with complete state and it is immediately ready, complete with network connectivity all in under 50ms.
Architect magic: Pod network migrations without pause and without custom CNI
Architect employs “dynamic real time routing” and doesn’t require custom CNI plugins or need some complex setup. It elegantly migrates connections using our eBPF network layer and maintains universal compatibility with any cluster.
! Let Your Cattle Sleep Standing Up
While you could manually put a pod to sleep, it is inefficient and requires manual intervention and modification of your code in order to capture state.
OLD WAY ~ MANUAL /
NEW WAY ~ WITH ARCHITECT /
╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
╎ Zzzzzzzzzzzzzzzzzzzz ╎
╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌
\ ^__^
\ (00)\_______
(__)\ )\/\
||----w |
|| ||
^^^-^--^^^-^^^^^^-^^^-^^/ Works the way you work
Architect works with all your standard Kubernetes configurations and cloud native technologies and you can try it today by following our quick start.
Save as you scale
Scale idle workloads to zero, gaining better bin-packing and cost savings.
Eliminate startup time
Pods start immediately as you need them. Even complex enterprise workloads.
Kubernetes native
Architect follows the same patterns and exposes primitives the way you already work.
> Get started: Run the way you run today
It’s easy to get started with Architect. It requires no code modification or complex analytics tooling. Just a helm install and labels and you are ready to rock.
Build your containers with standard tooling and deliver them without code modification.
Deploy with a simple helm chart on any infrastructure including spot instances, on nearly any cloud.
Hibernate or move your pods anywhere to optimize for utilization or cost, without interruption.
AAA RRRRRRRR CCCCCC HH HH IIIIII TTTTTTTT EEEEEEE CCCCCC TTTTTTTT AA AA RR RR CC CC HH HH II TT EE CC CC TT AA AA RR RR CC HH HH II TT EE CC TT AA AA RRRRRRRR CC HHHHHHHHH II TT EEEEEE CC TT AAAAAAAAA RR RR CC HH HH II TT EE CC TT AA AA RR RR CC CC HH HH II TT EE CC CC TT AA AA RR RR CCCCCC HH HH IIIIII TT EEEEEEE CCCCCC TT