Skip to content
ServerSpotter

How We Test

Transparency is core to what we do. Here's exactly how we benchmark, rate, and rank every cloud infrastructure provider on ServerSpotter.

Our Benchmarking Process

1

CPU Performance

We run sysbench CPU benchmarks (single-thread and multi-thread) on freshly provisioned instances. Results are measured in events per second across multiple runs to account for noisy neighbors.

2

Disk I/O

Sequential and random read/write speeds are tested with fio (4K and 1M block sizes). We measure IOPS and throughput on the default storage tier each provider ships.

3

Network Throughput

iperf3 tests measure bandwidth between datacenter regions and major internet exchange points. We also test latency to 5 global endpoints (US East, US West, EU, APAC, SA).

4

Provisioning Speed

Time from API request to SSH-accessible instance is measured across 5 deployments. Cold start vs warm start times are recorded separately.

5

True Cost Analysis

We calculate total cost including: base price, egress fees, snapshot storage, IP addresses, support tier requirements, and renewal pricing. No marketing-estimate figures.

6

Weekly Re-testing

All benchmarks are re-run on a weekly cycle. Pricing data is scraped daily. Historical data is preserved so you can track provider performance over time.

7

Uptime Monitoring

We monitor provider status pages and run independent health checks from 5 global locations. Reported uptime includes maintenance windows that providers publicly disclose.

8

Rating Algorithm

Our composite rating weights: Performance (35%), Price/Value (25%), Reliability (20%), Features (10%), Support Quality (10%). Community ratings are blended in when sufficient data exists.

Independence & Editorial Policy

ServerSpotter is editorially independent. No provider can pay for a higher ranking. Our ratings are algorithmic, based on benchmark data and community reviews.

We do earn affiliate commissions when you sign up through some links on our site. This never affects our rankings, ratings, or editorial content. We clearly mark affiliate links per FTC guidelines.

If a provider's performance drops, their rating drops — automatically, every week. We don't manually adjust scores.

Data Sources

  • Direct provisioning — We create real instances on each provider to run benchmarks.
  • Provider APIs — Pricing, plans, and specs are scraped from official APIs and documentation.
  • Community reviews — Verified user ratings contribute to our composite scores.
  • Status pages — Uptime data from provider status pages and our independent monitoring.

Questions about our methodology?

Learn more about us