English
Español
Français
Deutsch
Italiano
Português
العربية
Русский
日本語
The requirement
Evening peak traffic had pushed several 1G uplinks past 80 % for hours at a time. The ISP wanted 10G to the access racks, a 40G path toward the core, and 2.5G ports for subscribers who had already moved to Wi-Fi 6 routers. The rollout had to be incremental, because the budget was approved per rack rather than per year.
What was specified
| Model | Qty | Role in the project |
|---|---|---|
| RS1008K-8L-12S-10G-M3 | 8 | Access rack switch, 12 x 10G SFP+ uplinks, L3 routing |
| RS1024K-24S-2S-40G-M | 2 | Aggregation switch with 40G backbone port |
| RS1025K-4LG-2S-10G | 16 | 2.5G access for high-usage subscribers |
Why this split
The access model was chosen for its uplink count, not its port count: 12 SFP+ cages per rack means one rack can serve several downstream cabinets without a second aggregation hop. The 40G aggregation switch is the only place where a chassis-class port was justified. 2.5G was kept at the subscriber edge, where the actual demand was, instead of being pushed through the whole estate.
Result
- Uplink utilisation at peak dropped well below the saturation point after the first four racks were moved.
- Rollout proceeded rack by rack with the same model, so spares and firmware stayed uniform.
- The core link no longer had to be redesigned as each rack was upgraded.
Customer and site details are anonymised. Uplink and routing capacity must be verified against the operator's own traffic profile.
