For automated terminals over 2M TEU or conventional terminals over 3M TEU: Deploy 1 Scaled XD server. This will take the decking load off of XPS, but XPS is able to handle the decking requests should the Scaled XD client be offline.
For automated terminals over 4M TEU or conventional terminals over 6M TEU: Deploy 2 Scaled XD servers. This will spread the decking load across 2 Scaled XD instances to ensure consistent response times during bursts of heavy traffic and provides redundancy to protect the XPS server should one of the Scaled XD clients be offline.
The final number of N4 XD servers you deploy depends on the size of your operations and the rate of vessel discharge moves during peak hours.To determine if you need additional decking bandwidth, in the Decking History view in N4 you can review how much time is spent on each decking request. A rule of thumb would be that if and N4 XD server is using more than 40% of its bandwidth to handle decking requests, then additional N4 XD servers should be deployed.
For more information, see Configure the N4 XD server(s) (on page 1).
We do not recommend using a Scaled N4 XD service for conventional terminals with less than 2M TEU, since there is no measurable benefit to compensate for the added complexity, especially when XPS performance is fine.