Per-U colocation at IDACORE Boise starts at 1U with no power commitment minimum. In 2026, pricing runs 30–40% below comparable AWS or Azure deployments, driven by Idaho Power's commercial rates near $0.055/kWh — roughly half the national average — and flat, predictable billing with no egress surprises.
What's Actually Driving Your Cloud Bill Right Now?
If you've been running production workloads on AWS or Azure for more than a year, you already know the pattern. The compute line is manageable. It's the egress charges, the inter-AZ transfer fees, the storage I/O costs — the dozens of line items that don't show up in any estimate — that turn a $15,000/month projection into a $38,000 invoice.
That's not a hypothetical. A healthcare SaaS company we work with was running about $40,000/month on AWS before they moved their core infrastructure to IDACORE Boise. Their monthly bill dropped to $26,000. Same workload. Same performance requirements. The difference was eliminating egress fees on about 40TB of monthly data transfer and paying actual hardware costs instead of hyperscaler margins.
Per-U colocation doesn't make sense for every situation. If you need to spin up 200 servers for six weeks and tear them down, cloud is the right answer. But if you have steady-state infrastructure that's been running the same configuration for 12 months, you're almost certainly paying a significant premium for the flexibility you're not using.
What Does Per-U Colocation Actually Cost in Boise?
The honest answer is: it depends on power draw and connectivity, but the structure is straightforward.
At IDACORE Boise, you pay for rack space by the U, power by consumption, and bandwidth by the port or commit. That's it. No egress fees. No mysterious "data transfer" charges that appear mid-month. No surprise bills because someone ran a large backup job.
Idaho Power's commercial rates run around $0.055/kWh. For context, the national average for commercial power is closer to $0.12–$0.13/kWh. That difference compounds fast when you're running dense compute. A server drawing 500W continuously costs roughly $24/month in power here. In a Chicago or Northern Virginia facility, that same draw could cost $50–$55/month — before the facility marks up power delivery.
Here's a rough comparison for a common mid-market deployment: 10U of compute, 2U of storage, dual 10GbE connectivity, and about 20TB of monthly transfer.
| Cost Component | Hyperscaler (AWS equivalent) | IDACORE Boise (per-U colo) |
|---|---|---|
| Compute (12U equivalent) | ~$8,400/mo | ~$2,100/mo (hardware amortized) |
| Storage (20TB usable) | ~$2,300/mo | ~$400/mo (hardware amortized) |
| Bandwidth (20TB transfer) | ~$1,800/mo | Included in port commit |
| Power (est. 3kW draw) | Bundled in compute | ~$250/mo at Idaho rates |
| Support | $15K/yr Business Support | Included |
| Total (monthly) | ~$12,500 | ~$2,750 + hardware |
The hardware column is where people get tripped up. You own the servers in a colo model, which means upfront capital. A reasonable 12U compute and storage configuration might run $60,000–$80,000 in hardware. At a $9,750/month operating cost difference, that's a 6–8 month payback — and then you're pocketing the difference indefinitely.
Does Per-U Make Sense If You're Not Filling a Full Cabinet?
This is the question I hear most often from companies that assume colocation has a high floor. It doesn't — at least not here.
IDACORE Boise has a 1U minimum with no power commitment minimum. You can colocate a single server. That's not a teaser rate with hidden minimums; it's how we actually operate. We've had customers start with 2U of space and grow to multiple cabinets over two or three years.
The practical floor for per-U to make financial sense is roughly 4–6U of steady-state infrastructure. Below that, the management overhead of owning hardware starts to eat into the savings. Above that, the math almost always favors colo over cloud for workloads that don't need elastic scaling.
One scenario where per-U colocation makes immediate sense even at small scale: compliance requirements. If your workload is subject to HIPAA, PCI DSS, or NIST 800-53, you need a facility that's been audited against those frameworks. IDACORE Boise holds SOC 2 Type II, PCI DSS, NIST 800-53, SSAE-16, and HITRUST CSF certifications. Getting that coverage in a cloud environment requires purchasing compliance add-ons, dedicated instances, and often a third-party audit anyway. In a colo, the facility certification is part of what you're paying for.
What About Contract Terms and Carrier Access?
Most enterprise data center providers require 36-month terms. We do 12-month standard. That's not a promotional offer; it's how we structure deals because we'd rather earn the renewal than lock you in.
On the carrier side, IDACORE Boise has seven on-net providers: Zayo, Lumen/Level 3, Cogent, CenturyLink, Syringa, Cable One, and Hurricane Electric. We run our own ASN and have been doing BGP peering since before most cloud providers existed — including peering at the Seattle Internet Exchange. Having seven carriers on-net means you can multi-home, negotiate competitive transit pricing, and avoid the single-provider dependency that creates real outage risk.
Latency from Boise is 14ms to Salt Lake City, 22ms to Portland, and 23ms to Seattle. For most application workloads, that's effectively local. For disaster recovery from a Pacific Coast primary site, it's excellent — you're geographically separated from the seismic and wildfire exposure zones that affect coastal infrastructure, with round-trip times that support synchronous replication for most database workloads.
What Support Actually Looks Like
I'll be direct about something that doesn't show up in pricing tables: the support model matters more than most buyers account for when evaluating colo.
When you open a ticket with a hyperscaler, you get a queue. If you're on a standard support tier, that queue might be staffed offshore and working from a script. When something is actually broken at 2am, that experience is genuinely painful.
At IDACORE, when you call, you reach someone who knows your rack. Not someone reading a runbook. Someone who can walk to your cage, look at the physical hardware, and tell you what's actually happening. We've been running infrastructure in Boise for 30 years. The engineers who answer the phone are the same ones who built the network.
That's not a soft benefit. When a BGP session drops or a power circuit trips at 3am, the difference between a 15-minute resolution and a 3-hour ticket queue is a real business impact.
Frequently Asked Questions
What is the minimum commitment for per-U colocation at IDACORE Boise?
IDACORE Boise has a 1U minimum with no power commitment minimum. You can start with a single 1U server and scale from there. There's no requirement to lease a full cabinet or commit to a minimum power draw — which makes it one of the more accessible entry points for colocation in the Pacific Northwest.
How does IDACORE Boise colocation pricing compare to AWS or Azure in 2026?
On comparable workloads, IDACORE Boise runs 30–40% less than AWS, Azure, or GCP. A workload that costs $40,000/month on a hyperscaler typically lands around $24,000–$28,000 in a Boise colo environment. The biggest driver is eliminating egress fees and paying flat, predictable rates instead of per-GB transfer charges.
What certifications does IDACORE Boise hold?
IDACORE Boise is certified SOC 2 Type II, PCI DSS, NIST 800-53, SSAE-16, and HITRUST CSF. The facility is compliance-ready for HIPAA, financial services, and government workloads. These aren't self-assessments — they're third-party audited certifications, which matters for regulated industries that need to demonstrate due diligence.
What carriers are available at IDACORE Boise?
IDACORE Boise has seven on-net carriers: Zayo, Lumen/Level 3, Cogent, CenturyLink, Syringa, Cable One, and Hurricane Electric. Having seven carriers on-net means you're not dependent on a single provider for transit, and you can negotiate competitive bandwidth pricing directly. IDACORE operates its own ASN and has BGP peering experience going back 30 years.
Is IDACORE Boise a good option for disaster recovery from Seattle or Portland?
Yes. IDACORE Boise sits outside the Pacific Coast seismic and wildfire risk zones that affect Seattle and Portland infrastructure. Latency from Boise is 23ms to Seattle and 22ms to Portland — fast enough for most DR and replication workloads. The facility's geographic separation from the coast is a genuine risk diversification benefit, not just a marketing talking point.
If you're running steady-state infrastructure and your cloud bill has been climbing without a clear reason, the numbers on per-U colocation in Boise are worth running seriously — not as a thought experiment, but with your actual workload specs. We'll tell you honestly whether it makes sense for your situation. Get a real cost comparison from our infrastructure team and see what flat, predictable pricing actually looks like against your current invoice.
