Factories, banks, studios, and hospitals are all pushing more data between sites than last year. Metro spans that once cruised at 100G are now bursting. The next step is not just “go faster,” it is choosing the right optical lane and timing the jump from 400G coherent to 800G without blowing up rack space, power, or your route diversity. Use the guide below to pick lanes, size wavelengths, and price the move in a way finance will understand.
Plan around spans and headroom. 400G coherent is the sweet spot for most metro DCI today. 800G coherent is the next step for short, clean routes and high-density interconnects. Use the tools below to size waves, power, spectrum, and timelines without guesswork.
| Scenario | Typical choice | Why this works | Notes |
|---|---|---|---|
| Campus / short metro ring, clean fiber | 800G coherent pluggables or 2×400G | High density, fewer lambdas to manage | Keep spans short and OSNR healthy |
| Metro DCI under ~120 km | 400G ZR | Simplest ops, strong ecosystem | Use 2×400G for headroom or protection |
| Metro with amplifiers / mixed spans | 400G ZR+ / OpenZR+ | Extra reach and FEC options | Confirm line system compatibility |
| Regional multi-span | 400G ZR+ per lambda | More tolerant of distance | Consider lower-order modulations as needed |
- High density and simple operations
- Great for short metro DCI and campus rings
- Lower chassis cost, watch switch power budget
- Flexible modulation and reach profiles
- Line system integration and monitoring
- Useful for mixed spans and regional routes
| Item | Ask for | Why it matters |
|---|---|---|
| Route diversity | Entrance photos, first handhole IDs, crossing list | Avoid shared cuts on “redundant” paths |
| Optical tests | OTDR, light levels, OSNR snapshot | Baseline for future troubleshooting |
| Turn-up docs | As-built, channel plan, acceptance report | Fewer handoff surprises later |
| SLA details | MTTR language, maintenance windows | Predictable operations during upgrades |
- Gather traffic and growth, run the wavelength sizer
- Pull route photos and entrance proof
- Baseline OSNR and loss on spans
- Pick 400G or 800G per span
- Choose pluggables vs transponders
- Lock channel plan and protection
- Install, test, and capture acceptance files
- Run failover and maintenance drills
- Publish a short ops runbook
Use your metro spans, growth curve, and acceptance tests to choose lanes, not vendor hype. If the wavelength sizer and spectrum plan show that a small number of 800G lambdas will cleanly cover short, clean routes, schedule that step. If mixed spans or longer reach are in play, 400G coherent with room to add more lambdas is a proven path. Lock route diversity, capture turn-up evidence, and revisit the plan quarterly as traffic grows.
