Seattle companies choose IDACORE Boise for disaster recovery because it sits outside the Cascadia subduction zone, delivers 23ms round-trip latency to Seattle, and offers per-U colocation at $300/kW/month with no long-term contract. Idaho data residency and 12-month terms make it a fast, low-risk failover site.
Why Does Seismic Risk Matter for DR Site Selection?
If you're running production in Seattle, you already know about the Cascadia subduction zone. A magnitude 9.0 event isn't a hypothetical — it's a documented, overdue geological event that USGS models put at roughly a 1-in-3 chance in the next 50 years for the full rupture. Your primary site is in the blast radius. Your DR site can't be.
Boise sits well east of the fault line, outside the Cascadia impact zone and outside the wildfire corridors that have taken out Pacific Coast fiber and power infrastructure in recent years. This isn't marketing spin — it's basic geography. When you're picking a DR site, the entire point is decorrelated risk. A DR site 20 miles from your primary that shares the same fault line, the same grid, and the same wildfire season isn't a DR site. It's the same risk pool with extra latency.
We built IDACORE Boise in Idaho specifically because it gives Pacific Northwest companies real geographic separation without giving up the latency they need for active-active or warm-standby architectures.
What Does DR Actually Cost at Boise Compared to Staying On the Coast?
Most DR conversations get derailed by vague pricing. Here's what it actually looks like.
Say you're running a 10kW footprint for DR — enough for a couple dozen servers handling database replication, backup storage, and failover application instances. At IDACORE Boise, that's $3,000/month flat, billed on power draw, not rack space. No minimum commitment. No 36-month contract forcing you to guess your DR needs three years out.
| Factor | Boise DR Colocation | Typical Seattle-Metro Colocation | AWS us-west-2 DR (comparable) |
|---|---|---|---|
| 10kW monthly cost | $3,000 flat | $3,500–$4,500+ | Variable, often $4,000–$6,000+ with egress |
| Contract term | 12 months | 36 months typical | Monthly, but egress unpredictable |
| Seismic zone | Outside Cascadia | Inside Cascadia | Inside Cascadia (same region as primary) |
| Latency from Seattle | 23ms | <5ms | Varies |
| Egress billing | Flat, no surprises | Varies | Per-GB, scales with recovery traffic |
Notice the AWS row. If your DR site is in the same seismic zone as your primary, you haven't actually solved the problem you set out to solve. That's the mistake we see most often — companies replicate to another AWS region that's still sitting on the same fault system, and call it disaster recovery.
What About Recovery Time and Bandwidth During Failover?
A DR site only matters if you can get to it — and get your data out of it — when you need to. This is where egress billing quietly wrecks DR budgets. During an actual failover event, you're pushing far more traffic than baseline: full database syncs, application state transfer, client redirect traffic. Hyperscaler egress bills spike exactly when you need the site most, and that spike is unpredictable by design.
Our IP transit is priced on committed bandwidth, not consumption. A 10G commit runs $1,350/month flat, billed at 95th percentile — whether you're pushing 2Gbps on a quiet Tuesday or maxing out during a failover event triggered by a real incident. We run this transit ourselves, off our own ASN, peered directly at facilities across the Pacific Northwest. There's no reseller markup, and no surprise invoice three weeks after your DR test.
How Fast Is Failover From Seattle to Boise, In Practice?
23ms round-trip from Boise to Seattle is fast enough for synchronous-adjacent replication on most database engines, and comfortably fast for asynchronous replication with sub-second lag under normal load. For comparison: 22ms to Portland, 14ms to Salt Lake City. If your architecture already tolerates cross-AZ latency within a single AWS region (often 1-10ms but sometimes higher across AZs in congested regions), 23ms to a fully separate seismic zone is a reasonable tradeoff for real geographic diversity.
A healthcare SaaS company we work with runs primary infrastructure in Seattle and replicates PHI-adjacent workloads to Boise for HIPAA-aligned DR. Their compliance team required data residency within the Pacific Northwest — Idaho qualified, an out-of-region cloud DR site didn't fit their audit story as cleanly. They're running a 15kW footprint at $4,500/month, replacing what was a $9,200/month multi-region AWS setup with cross-region data transfer fees baked in.
What Compliance Coverage Does Boise Provide for DR Workloads?
If your primary workload runs under HIPAA, PCI DSS, or SOC 2 obligations, your DR site needs equivalent controls — not "close enough." A DR environment that can't pass the same audit as your primary defeats the purpose of having one.
IDACORE Boise is certified SOC 2 Type II, PCI DSS, NIST 800-53, SSAE-16, and HITRUST CSF. That's not a claim we make loosely — it's audited coverage that lets your compliance team sign off on Boise as a DR site without a separate risk exception process. For regulated workloads, that paperwork matters as much as the infrastructure itself.
Does 12-Month Term Length Actually Matter for DR?
Yes, more than people expect going in. DR needs change. You might start with a 5kW cold-standby footprint and need 20kW within a year once you've proven out active-active replication. Enterprise colocation providers typically lock you into 36-month terms, which means your DR footprint is either oversized on day one (paying for capacity you don't need yet) or you're stuck renegotiating a multi-year contract mid-term.
Our 12-month standard term at Boise lets you scale DR capacity in step with actual recovery testing results, not a three-year forecast made before you'd run a single failover drill.
Frequently Asked Questions
Is Boise far enough from Seattle to be a real disaster recovery site?
Yes. Boise sits outside the Cascadia subduction zone, the primary seismic threat to Seattle infrastructure, while maintaining 23ms latency for practical replication. That combination — real geographic separation plus workable latency — is harder to find than it sounds, since many "regional" DR sites still share the same fault system as the primary.
How much does DR colocation cost at IDACORE Boise?
Colocation is billed at $300/kW/month based on power draw, not rack space. A 10kW DR footprint runs $3,000/month flat with no minimum power commitment and 12-month standard terms, versus 36-month terms typical of enterprise colocation providers.
Can I replicate HIPAA or PCI workloads to Boise for compliance-aligned DR?
Yes. IDACORE Boise is certified SOC 2 Type II, PCI DSS, NIST 800-53, SSAE-16, and HITRUST CSF, with compliance-ready infrastructure for HIPAA and financial workloads. This lets regulated companies run DR in Idaho without a separate audit exception for the failover site.
What's the latency between Seattle and Boise for database replication?
23ms round-trip. That's workable for asynchronous replication on most database engines and often comparable to cross-AZ latency within a single congested cloud region, while providing actual seismic and wildfire risk separation that same-region cloud DR doesn't.
Does IDACORE charge egress fees during a DR failover event?
No. IP transit is priced on committed bandwidth tiers billed at 95th percentile — for example, $1,350/month for a 10G commit — not per-GB egress. That means failover traffic spikes don't generate surprise bills the way hyperscaler egress billing does.
If you're planning a DR site for 2026 and want geographic separation from Cascadia without giving up latency to Seattle, talk to the team that built and runs Boise's network ourselves — get a DR footprint quote at IDACORE.
