Skip to main content
Version: 2.0.0 (Latest)

System requirements

This page lists what users and admins need to run PADAS in production or staging environments.

PADAS runs as a standalone service. External systems such as Kafka, Splunk, syslog devices, and object storage are optional and only required when you enable matching connectors.

Operating system

ItemRequirement
Primary targetLinux x86_64
Service managersystemd recommended
Container optionTBD (not documented in this site yet)
ComponentRecommended baseline
CPU4+ cores
Memory8 GB minimum, 16 GB preferred
DiskSSD; size based on logs + retention + WAL usage

Optional integrations (via connectors)

Connectors are how PADAS talks to the outside world. Each is optional — pick the ones your deployment needs. Source connectors push events into streams; sink connectors drain them out.

ConnectorDirectionWhen to use
PADAS TCPsource / sinkEngine-to-engine streaming over the native PADAS wire format. No external dependencies.
HTTPsource / sinkPush or pull over HTTP APIs.
Filesource / sinkNDJSON or text files. The simplest local-development connector — no extra infrastructure.
Syslogsource / sinkRFC 3164 / 5424 over UDP/TCP. Requires reachable network interfaces.
Splunk HECsource / sinkHTTP Event Collector compatibility — needs a reachable Splunk instance (or HEC mock for tests).
Kafkasource / sinkUse when you already operate Kafka. Requires reachable broker(s); KRaft-mode Kafka does not need a separate ZooKeeper. Not required for PADAS to run.
Object storage (S3 API)sinkS3-compatible storage with Parquet or JSON Lines for archive flows.

Network and access

ItemRequirement
Inbound accessEngine HTTPS API on 8999 (paths under https://<CORE_IP>:8999/api/v1/… unless changed); health (/health or equivalent) reachable from operators and gateways
Outbound accessReachable endpoints for enabled connectors (Kafka, Splunk, S3, etc.)
Service accountNon-root runtime user with write access to data and log directories

Where to next