Jump Host

A jump host (jump server / bastion) is an architecture where remote users connect via RAS or VPN not to the whole control network, but to a specific workstation or host. They typically receive keyboard, video and mouse (KVM-style) access and work with tools installed on that host — they “jump” onto it.

Teaching note: Paraphrased for learning from IACS cybersecurity course material and industry practice. Not a substitute for ISA/IEC 62443 remote-access requirements (Part 2-1 NET 3) or organisation-specific architecture standards.

How it works

Users authenticate to a remote-access path (for example a VPN or remote-access service) that lands them on the jump host rather than granting membership of the broader IACS network. From the jump host they use approved applications to reach selected engineering or operations systems.

Benefits

You negotiate which programs remote users need, while retaining control of what is installed on the host
You can limit which systems or assets the jump host can reach
You can block file transfers into and out of the control system

Placement and hardening

Place jump hosts in the IACS DMZ (or equivalent restricted zone) and grant them access only to selected resources in the control system — not blanket network membership. Harden the jump host as a high-value endpoint (patching, least privilege, MFA, logging) and treat it as part of secure remote access design, not as a convenience PC.

Key takeaways

A jump host concentrates interactive remote work on a controlled host instead of exposing the control network.
Limit reachability, installed tools and file transfer to shrink blast radius.
Prefer DMZ placement with MFA, monitoring and hardening aligned to the zone’s SL-T.