System Requirements

System requirements for jadice flow's components are given below. For an overview of the components involved, please consult images and description in the system architecture section. All components are packaged as Linux amd64 (x86-64) Container Images.

Required specs:

• 2 CPU and 4 GB RAM for the jadice flow controller
• 2 CPU and 2 GB RAM for the controller UI (vaadin based UI might use more memory under heavy user load, which is not expected in the general use case)
• 2 CPU and 4 GB RAM for each worker pod in general - some workers like optical character recognition or certain image conversions might utilize a lot CPU resources depending on the input documents (also required in single core CPU area). Check the worker description and the table below if a specific recommendation is available.

Client

• Either REST API:
  • No special requirements.
• or Flow Java Client:
  • Java VM Version 11
  • jadice flow controller-client for server communication

Server

Container orchestration:

• Kubernetes, OpenShift (recommended)
• or Docker Compose, Swarm (for development purposes only)

When performing a Kubernetes deployment using the Helm Charts delivered with the products, the resources are already configured as required.

For reference purposes, we still list the resource requirements for containers here:

component	cpu	memory	persistent storage
controller	0.5	2 GB	-
controller-ui	0.5	2 GB	-
controller database	0.1	256 MB	min 1 GB, depending on throughput and retention period

Helm Container Platform Capacity Resource Requirements depending on the desired conversion:

worker	Pod Minimum CPU	Pod Minimum Memory
to-TIFF	0.25	512 MB
to-PDF	0.25	512 MB
OCR	1	2 GB
imagemagick	1	2 GB
libreoffice	1	2 GB
analyzer	0.2	256 MB
decompress	0.2	256 MB
email-disassembly	0.2	256 MB
HTML-to-PDF	0.2	256 MB
dynamic-controller	0.2	256 MB
mail-body-creator	0.2	256 MB