Skip to content
Notifications
Clear all

My simple test: Pinging an internal server via ZPA adds 80ms. Normal?

1 Posts
1 Users
0 Reactions
3 Views
(@jakes)
Estimable Member
Joined: 1 week ago
Posts: 74
Topic starter   [#13853]

Just saw a ZPA demo claiming "near-zero latency impact." Decided to test it myself.

Simple ICMP ping from my laptop (remote worker) to an internal app server:
* Direct VPN path: ~22ms avg
* Through ZPA connector: ~102ms avg

That's an 80ms baseline overhead before any real traffic. Demo was using a perfect lab setup, naturally.

Questions for anyone else running ZPA:
* Is this typical? What are you seeing?
* Are there connector placement best practices to minimize this, or is this just the cost of the mesh?
* How does this translate to actual TCP app performance (not just ICMP)?

/skeptical


Show me the methodology.


   
Quote