Reviewed measurements

Compare the evidence.

Two reports. The same workloads. Measured medians in view.

Selected plans
DigitalOceans-4vcpu-8gb
Vultrvoc-c-4c-8gb-75s-amd
Different resources selected: 4 vCPU · 7.8 GiB / 4 vCPU · 7.7 GiBSource location: Bangalore, IN · same agent build and methodology.
CPU allocation
4 vCPU
4 vCPU
Observed memory
7.8 GiB
7.7 GiB
Processor
DO-Regular
AMD EPYC-Rome
Location
Bangalore, INblr1
Bangalore, INBLR
OS / architecture
Ubuntu 24.04.5 LTS / x86_64
Ubuntu 24.04.5 LTS / x86_64
Sampling
1 deployment · 24h run skipped
1 deployment · 24h run skipped
Report date
17 Oct 2026
17 Oct 2026
Methodology
pilot-1 · AI-assisted review
pilot-1 · AI-assisted review
Each column is one deployment. Matching context does not establish equivalent physical hosts.Reviewed data · within-run medians
PerformanceReviewed data
DigitalOcean4 vCPU · 7.8 GiB
Vultr4 vCPU · 7.7 GiB
01 /

CPU performance

sysbench prime events · max prime 20,000 · gzip level 6 · higher is better

MetricDigitalOcean s-4vcpu-8gbVultr voc-c-4c-8gb-75s-amd
Single thread
360.3events/s24% of 1,500
1,333events/s89% of 1,500
All 4 vCPUs
1,444events/s24% of 6,000
5,299events/s88% of 6,000
gzip compression
160.3MiB/s80% of 200
160.9MiB/s80% of 200

Identical thread counts and test settings. Bars use a common scale within each row.

02 /

Memory bandwidth

sysbench · 256 MiB block · one thread · higher is better

MetricDigitalOcean s-4vcpu-8gbVultr voc-c-4c-8gb-75s-amd
Read
5,872MiB/s39% of 15,000
13,023MiB/s87% of 15,000
Write
6,124MiB/s41% of 15,000
12,513MiB/s83% of 15,000

Bars share one scale across both rows. A memory copy workload is not part of this profile.

PerformanceReviewed data
DigitalOceanBangalore, IN
VultrBangalore, IN
03 /

Disk performance

fio · direct I/O · 1 MiB / QD32 sequential · 4 KiB / QD32 random

MetricDigitalOcean s-4vcpu-8gbVultr voc-c-4c-8gb-75s-amd
Sequential read
4,585MB/s
3,421MB/s
Sequential write
2,212MB/s
2,410MB/s
Random read
84,421IOPS
92,339IOPS
Random write
61,228IOPS
79,365IOPS
Random read p95
585.7µs
473.1µs
Random write p95
880.6µs
593.9µs

Throughput and IOPS: higher is better. Latency: lower is better. Mixed 70/30 runs and p50/p99 latency are in each full report.

04 /

Network performance

Same two controlled London endpoints · one iperf3 stream · 100 Mbit/s rate setting

MetricDigitalOcean s-4vcpu-8gbVultr voc-c-4c-8gb-75s-amd
Download · endpoint 1
99.9Mbit/s
92.3Mbit/s
Download · endpoint 2
14.7Mbit/s
99.9Mbit/s
Upload · endpoint 1
97.5Mbit/s
97.4Mbit/s
Upload · endpoint 2
29.5Mbit/s
95.5Mbit/s
Latency · endpoint 1
131.2ms
152.7ms
Latency · endpoint 2
163.5ms
247.7ms

Source location affects latency. Actual bytes can exceed the 64 MiB target; these routes do not measure maximum link capacity. Jitter and packet loss are not reported.

PerformanceReviewed data
DigitalOcean4 vCPU · 7.8 GiB
Vultr4 vCPU · 7.7 GiB
05 /

PostgreSQL performance

PostgreSQL 16 · pgbench read/write · 4 clients · durability enabled

MetricDigitalOcean s-4vcpu-8gbVultr voc-c-4c-8gb-75s-amd
In-memory throughput
2,807transactions/s
2,670transactions/s
In-memory average latency
1.42ms
1.50ms
Storage-bound throughput
2,459transactions/s
1,920transactions/s
Storage-bound average latency
1.62ms
2.08ms

Both dataset profiles shown. This profile records average latency, not percentiles; read throughput and latency together.

06 /

Consistency over 24 hours

Separate 24-hour profile · hourly CPU, memory read and random disk read

Hourly repeat runsCPU · memory · disk
DigitalOcean
Not measured24-hour run skipped
Vultr
Not measured24-hour run skipped
No hourly samples exist for these reports, so no chart is drawn.Skipped runs are not scores.

No 24-hour consistency or long-term reliability is established.

07 /

Inspect the source reports.

Different workloads reveal different trade-offs. Every number links back to its test.