Every VPS provider advertises an uptime number, and almost every one of them sounds reassuring at a glance: 99%, 99.9%, sometimes 99.99%. The problem is that these numbers compress into a tiny visual difference on a marketing page while representing enormous differences in actual downtime hours. For a trader whose EA needs to be connected the moment a stop-loss needs adjusting or an entry signal fires, that difference is not academic.
Here is what these numbers actually mean in hours and minutes, what an SLA does and does not cover, and how to evaluate one honestly.
The Math Behind the Percentages
| Advertised Uptime | Downtime Per Year | Downtime Per Month |
|---|---|---|
| 99% | ~3.65 days | ~7.3 hours |
| 99.5% | ~1.83 days | ~3.6 hours |
| 99.9% | ~8.76 hours | ~43.8 minutes |
| 99.95% | ~4.38 hours | ~21.9 minutes |
| 99.99% | ~52.6 minutes | ~4.4 minutes |
📊 Key Stat: The jump from 99% to 99.9% looks small on paper (0.9 percentage points) but represents the difference between roughly 3.65 days of downtime per year and roughly 8.76 hours. The jump from 99.9% to 99.99% represents the difference between 8.76 hours and under an hour. Small-looking percentage differences hide large absolute differences at this end of the scale.
For a trader, the relevant question is not “is 99% close to 100%?” It is “what happens to my open EA positions during however many hours of downtime that percentage implies?” Three and a half days of unplanned downtime spread across a year, landing at an unpredictable moment, is a materially different risk than under an hour.
What FXVPS Actually Commits To
FXVPS datacenters, including London (LD4), New York (NY4), Tokyo (TY3), and Hong Kong (HK1), operate at 99.99% uptime, backed by redundant power, battery backup, and diesel generator failover. In practical terms, that is the difference between an EA that occasionally survives a brief, well-managed interruption, and one that is genuinely exposed to multi-day outages spread unpredictably across the year.
What an SLA Actually Covers (and What It Doesn’t)
An uptime guarantee is not a blanket promise that nothing will ever go wrong. It is a specific commitment about the infrastructure layer, and it is worth understanding where that commitment starts and stops.
Typically covered by a provider’s SLA:
- Physical server and hypervisor availability
- Network connectivity within the datacenter
- Power delivery to the physical hardware
Typically NOT covered by a provider’s SLA:
- Your operating system crashing or needing a Windows Update reboot
- Your trading terminal freezing or your EA crashing
- Your broker’s own server being down or your broker suspending trading
- Scheduled maintenance windows the provider disclosed in advance
⚠️ Warning: A provider’s 99.99% infrastructure uptime does not automatically mean your EA had 99.99% uptime. If MT4 crashed due to a bad indicator, or a Windows Update rebooted the machine without auto-start configured, that downtime happened on infrastructure that was technically “up” the entire time. The SLA covers the datacenter’s responsibility, not yours. See our guide to automating EA restarts for closing that gap.
Questions Worth Asking Before You Trust a Number
💡 Tip: Ask a provider these four questions before trusting their advertised uptime figure. A provider confident in its infrastructure will answer all four directly and specifically.
-
Is the number backed by an actual SLA with defined remedies, or is it just marketing copy? A real SLA specifies what happens (credits, refunds) if the provider misses the target. A number with no enforcement mechanism behind it is a claim, not a guarantee.
-
Does the datacenter have redundant power? Battery backup alone covers short outages. Diesel generator failover is what covers an extended utility power loss without the servers going dark.
-
Is network connectivity redundant, or is there a single point of failure? Multiple upstream network providers with automatic failover protect against an ISP-level outage taking the whole facility offline.
-
What counts as a “scheduled maintenance” exclusion, and how much notice do you get? Some providers exclude planned maintenance from their uptime calculation entirely, which can quietly erode the real-world number if maintenance windows are frequent or long.
Reading Between the Lines of Marketing Language
✅ Best Practice: Treat “up to 99.9% uptime” and “99.9% uptime” as different claims. “Up to” is a ceiling, not a commitment — it describes the best case, not what you should plan around. A provider stating a flat, unqualified percentage backed by an actual SLA document is making a materially stronger claim.
Similarly, “enterprise-grade” and “redundant” are frequently used without specifics. Ask what redundancy actually means at the facility: N+1 power? Dual utility feeds? Multiple carrier-diverse network paths? The specific architecture behind the claim tells you far more than the adjective describing it.
Why This Matters More for Traders Than for Typical Web Hosting
A website going down for ten minutes is an inconvenience. An EA going offline for ten minutes during a fast-moving session, with an open position that needed a trailing stop adjustment or a scheduled close, is a direct financial cost. The bar for acceptable uptime is genuinely higher for trading infrastructure than for most other VPS use cases, which is why datacenter selection and provider transparency matter more here than in general-purpose hosting decisions.
Frequently Asked Questions
Is 99.9% uptime good enough for forex trading?
It depends on how that downtime is distributed and what your strategy tolerates. 99.9% still allows nearly 9 hours of downtime per year, and if even one of those hours lands during an open, unmanaged position, the cost can outweigh months of saved hosting fees. Higher-tier commitments (99.95%+, ideally 99.99%) meaningfully reduce that exposure.
Does FXVPS’s uptime commitment include my EA and terminal staying connected?
FXVPS’s 99.99% figure covers the datacenter infrastructure: power, network, and hypervisor availability. Keeping your specific terminal and EA running continuously on top of that requires the auto-start and watchdog automation covered in our Task Scheduler guide, since terminal-level crashes are outside any hosting provider’s infrastructure SLA.
What’s the difference between uptime and latency?
Uptime measures whether the infrastructure is available at all. Latency measures how fast an available connection responds. A VPS can have perfect uptime and still deliver poor execution if it is not colocated near your broker’s matching engine — see our latency and slippage guide for that side of the equation.
How do I verify a provider’s uptime claim independently?
Look for third-party status pages, historical incident reports, or reviews that reference specific outages rather than general satisfaction. A provider with nothing to hide typically publishes a public status history rather than only stating a percentage on its marketing page.
Related Reading
- Forex VPS Performance Metrics: What to Track for Success — monitoring your own uptime on top of the provider’s infrastructure
- Automating EA Restarts and Reboots with Task Scheduler — closing the gap between infrastructure uptime and terminal uptime
- Important Differences Between VPS Hosting and Dedicated Hosting — how infrastructure architecture affects reliability
- Running Expert Advisors 24/7: Complete Guide — the broader case for always-on infrastructure