Sizing
Weik.io 2026.2Size the hub and the agents from the measured image size and memory of each part of Weik.io.
The image sizes, and the memory of the integrations and of the hub as a whole, come from measurements on the built images and the tested installations. For the first three parts in the table below, the memory column has the range each part is designed to run in.
Once the installation runs, the Health page shows the resources each part uses. The problem list shows a problem when an agent’s host has more than 90 % of its memory in use or less than a fifth of its disk free.
What runs where#
| Part | Where | Image | Memory |
|---|---|---|---|
| Message server | The hub, on each server | About 26 MB | 40 to 150 MB |
| Backend, which serves the admin UI and the API and builds versions | The hub, on each server | About 450 MB, because it contains a Java runtime and a Camel launcher for each supported Camel version | 130 to 300 MB, more while it builds a version |
| API gateway | The hub, on each server, and a gateway at a site | About 140 MB, and about 400 MB more for keeping a site gateway’s telemetry | 90 to 200 MB, and 200 to 230 MiB more at a site for its telemetry |
| Log, metric and trace stores | The hub, on node 1 of several servers | One binary each | About 20, 70 and 15 MB idle, up to about 150, 280 and 110 MB after an hour of eight busy integrations |
| Agent | Each agent’s host | About 430 MB, and about 400 MB more for keeping the agent’s telemetry | About 110 MB, and 200 to 230 MiB for the agent’s telemetry |
| Leaf message server | Each agent’s host | 26 MB | 26 MB |
| One Camel integration | Each agent’s host, one JVM per running deployment | Built at the hub | 136 MiB with traces and metrics on, 131 MiB with them off |
Eight integrations on one agent use 820 to 880 MB together, less than eight times one, because they share the platform’s classes. Each integration is ready about 0.38 seconds after it starts.
A hub on one server#
A hub needs at least 2 GB of memory. On a host with less, init warns about it. The services and the three stores use about 500 MB when idle on the tested hub, a server with 4 virtual CPUs and 8 GB of memory that runs Ubuntu 24.04 LTS on x86-64.
A developer stack with the hub and an agent on one machine stays under 2 GB with up to five integrations.
While weikio apply builds a version on the hub, each Java process of the build uses at most 256m of heap. You can set a size from 128m to 16g with weikio settings set --build-heap 512m. Settings › Platform › Installation shows the size as Build memory.
| For | Disk space |
|---|---|
| The message server’s storage | One and a quarter times the storage budget |
| The images | About 2 GB |
| The telemetry stores | Up to about 75 GB with their default caps |
The tested disk of 160 GB has room for all of them with --disk-budget 40GiB and about 29 GB to spare.
Storage and its budget has the disk each telemetry store uses and how to change the caps and the days.
A hub on several servers#
| Role | Each server |
|---|---|
| Data server, nodes 1 to 3 or 1 to 5 | 4 virtual CPUs, 8 GB of memory and a disk for its copy of the storage |
| Standard server | 4 virtual CPUs, 8 GB of memory and about 10 GB of disk |
| Node 1, in addition | Disk for the log, metric and trace stores, up to about 75 GB with their default caps |
For example, a data server’s disk of 100 GB under the message store fits --disk-budget 74GiB. That budget has room for three environments and 26 projects of the default size beside the installation’s stores.
Events a second#
The speed of the disk under the hub’s storage sets how many events and other changes the hub accepts a second. The tested hub had three servers of 4 virtual CPUs and 8 GB of memory, each with a copy of the storage.
| Disk under the storage | Time of one write | Events the hub stored |
|---|---|---|
| The server’s own solid-state disk | about 1 ms | about 600 a second |
| A network-attached disk, encrypted or not | 5 to 9 ms | 70 to 100 a second |
Measure the disk once it’s mounted. The command prints the time of 1,000 writes, so divide it by 1,000 for the time of one write:
$ dd if=/dev/zero of=/data/disk-speed bs=4k count=1000 oflag=dsyncPut the hub’s storage on the fastest disk your encryption rules allow. A dedicated server with a second local disk gives the rate of a server’s own disk.
An agent#
| For | Memory |
|---|---|
| The agent | About 110 MB |
| Each integration it runs | 131 to 136 MiB |
| The telemetry the agent keeps while the hub is out of reach | 200 to 230 MiB |
Each integration also needs room for its Java heap, which may grow to 256 MB unless you give its deployment another size.
| For | Disk space |
|---|---|
| The images | About 0.9 GB, and about 0.55 GB more for a gateway’s |
| The agent’s telemetry | Up to 4 GiB |