Choosing Between Two Metros for Colocation or a Second Site

When this decision comes up

Buyers usually end up comparing two metros in one of three situations: picking a first colocation site where two cities are plausible, adding a second production site for growth or latency, or choosing a disaster recovery location. The scoring below works for all three; the weights change.

If the second site exists mainly for recovery, read Disaster Recovery Colocation first, because recovery objectives drive the market choice. If you are still deciding on providers, How to Choose a Colocation Provider covers the facility-level criteria.

The seven factors to score

1. Latency and user location

Where are your users, customers, offices and cloud regions? A metro closer to where traffic originates or terminates reduces latency for every request. If the new site must replicate synchronously with an existing one, distance between the two sites becomes a hard constraint. Ask your network and storage teams what latency your design tolerates, then rule out markets that cannot meet it.

2. Separation from your other site

For a second site, the two metros should not share the same likely failure causes. Consider weather and natural hazards, the electric utility and transmission region, and the major fiber routes in and out. Two cities in the same storm track or on the same utility grid offer less protection than their distance suggests.

3. Power and space availability

Power is often the gating resource. Some large markets have strong demand and limited near-term capacity in certain buildings, while others have more room to grow. Ask each provider whether the quoted capacity is live or depends on future construction, and whether you can reserve expansion. Colocation Power Requirements explains how to state your power needs so quotes are comparable.

4. Carriers, clouds and interconnection

Count what is on-net in the specific buildings you would use, not the metro overall. A carrier-neutral data center in a major hub may give you more carrier and cloud choice than a single-network facility in a smaller market. If hybrid cloud matters, see our post on cloud on-ramps in colocation.

5. Staff access and operations

Who will go on site, and how often? A market near your team, or with a major airport and easy driving access, makes installs and incidents simpler. If no one will be local, weigh the quality and scope of remote hands in each candidate building more heavily.

6. Provider footprint

If one provider operates in both metros, you may gain consistent contracts, a single portal and private links between sites. Using different providers in each metro reduces dependence on one company. Neither is always right; decide which risk you would rather carry.

7. Total quoted cost

Market rates differ by power availability, demand, taxes and utility costs. Rather than relying on published averages, request quotes on the same requirements in both metros and compare the totals, including one-time costs and connectivity. How to Compare Colocation Quotes shows how to normalize them, including across markets.

A simple scoring sheet

Weight each factor for your situation, score each metro, and multiply. Keep the weights honest; the point is to make trade-offs visible.

For a disaster recovery site, separation and network diversity usually carry the most weight. For a growth or latency site, user proximity and interconnection usually do. For a first site, total cost and power availability often decide it.

Common metro pairings to evaluate

These are starting points for your own analysis, not recommendations. Validate hazard exposure, carrier routes and capacity for your design.

• Dallas or Phoenix. Both are large, growing markets with many providers. Buyers often compare Dallas, a central interconnection hub, with Phoenix, a common choice for West Coast companies looking inland.
• Atlanta or Chicago. Atlanta is the main Southeast hub, and Chicago is the main Midwest hub. Companies serving the eastern half of the country often weigh one against the other, or use both.
• Denver or Salt Lake City. Denver and Salt Lake City are frequently considered as inland secondary sites for organizations with a coastal primary.
• Kansas City as a central option. Kansas City comes up for buyers who want a central location away from the largest hubs.

Browse every market on the data center directory to see which providers operate in each.

How to gather the data for each factor

Scoring only works if the inputs are real. A practical way to collect them:

• Latency: ask your network team to measure round-trip times from user locations, offices and cloud regions to test points in each metro, or ask shortlisted providers for test IPs.
• Separation: compare each metro's utility territory and the main fiber routes your carriers use, and check public hazard maps for weather and seismic exposure.
• Power and space: ask each provider whether the quoted capacity is live, reserved or planned, and what expansion can be held for you in writing.
• Carriers and clouds: request building-level carrier and on-ramp lists, not metro marketing sheets.
• Staff access: map travel from where your people actually are, including how quickly someone could reach each site during an incident.
• Cost: send the same requirements document to providers in both metros, with the same term and redundancy, and normalize the responses before comparing.

If a factor cannot be measured yet, score it conservatively and mark it as an assumption so it gets revisited once quotes arrive.

Mistakes to avoid

• Choosing a market on reputation without quotes from both.
• Comparing metro-level carrier lists instead of building-level ones.
• Picking a second site that shares the primary's utility region or hazard profile.
• Ignoring staff travel for installs and incidents.
• Assuming capacity is available because the market is large.

Next step

The fastest way to decide between two metros is to put real quotes from both side by side. Request quotes in both markets, or book a free consult to build your scoring sheet with someone who compares markets every day.

Frequently asked questions