Akamai (Linode) g6-dedicated-4
Network-only report. Other performance workloads were not measured in this run.
One download and upload trial per named destination; missing outcomes are not scores.
Network performance
Source: London 2, UK · LibreSpeed CLI 1.0.14 · public HTTPS destinations.
| Destination | Download | Upload | HTTP latency | HTTP jitter | Loss |
|---|---|---|---|---|---|
| Frankfurt, Germany (Clouvider) 18/10/2026, 12:07:25 UTC · Same deployment supplement | 4,164 Mbit/s | 1,195 Mbit/s | 14.1 ms | 0.20 ms | Not measured |
| New York, United States (2) (Clouvider) 18/10/2026, 12:07:25 UTC · Same deployment supplement | 1,077 Mbit/s | 308.5 Mbit/s | 70.9 ms | 0.14 ms | Not measured |
One 10-second download and upload per destination, three concurrent streams, IPv4 over HTTPS, with no configured rate limit. HTTP latency and jitter are application-level measurements; packet loss was not measured. These paths do not establish maximum link capacity.
Same deployment as the original full benchmark, recorded in a separate test window. Environment observations are inherited from that linked report.
Outbound traffic has a monitored byte-abort threshold with sampling and termination overshoot; an aborted transfer is unscored.
Inspect the evidence.
Recorded environment, reviewed measurements and the published snapshot.
Reviewed, sanitized snapshot: every run, the environment record and all limitations. Raw intake logs remain private.
One deployment describes this report, not every server a provider operates.
A 49 seconds test window does not establish 24-hour consistency or long-term reliability.
AI-assisted review · AI-assisted review · Reviewed 18 Oct 2026 · Published 18 Oct 2026 · Revision 1
Review and measurement limits
- Network-only supplement; not a complete performance run and not a replacement CPU/disk/database result.
- Speeds measure throughput to the named public test servers, not guaranteed port capacity or every Internet destination.
- One 10-second download and upload per target, three concurrent streams; independent VM replicas remain separate.
- HTTP latency and HTTP jitter are application-level observations, not ICMP RTT or packet loss. Packet loss was not measured.
- Fixed comparison destinations: Frankfurt, Germany (Clouvider), New York, United States (2) (Clouvider). These are public Internet paths, not provider-local capacity tests.
- Environment observations come from the linked original report on the same deployment.
- Public endpoint load and HTTP error behavior can constrain rates. Raw stdout/stderr are retained; a positive rate is not proof of endpoint capacity.
- Outbound traffic has a monitored byte-abort threshold with sampling and termination overshoot; an aborted transfer is unscored.
- Short profile: does not establish 24-hour consistency.
- This report received AI-assisted evidence review under user-delegated authorization. This is not independent human verification or hardware attestation.
- This pilot measures its recorded test window. It does not establish 24-hour consistency, long-term reliability or provider-wide performance.
- Provider, region and SKU were matched to retained deployment records. Missing detected values use operator-supplied context and remain labelled supplied. Control-plane records and checksums do not attest the physical host or establish physical-host independence.
Environment and configuration
| Recorded field | Value |
|---|---|
| Provider | Akamai (Linode) · detected |
| Region | London 2, UK (gb-lon) · detected |
| Plan | g6-dedicated-4 · detected |
| CPU | AMD EPYC 7713 64-Core Processor |
| Architecture | x86_64 |
| Kernel | 6.8.0-134-generic |
| Virtualization | kvm |
| Agent | 0.2.0 (network-supplement-v1+861ada718c831d572d880d407f046e4e917c1672) |
| Dependencies | librespeed-binary-sha256: aff3064af531915e79dd97aa35e36e4efad5a2dd89e22cea2b2263cadeb63317 · librespeed-cli: 1.0.14 |
| Configuration hash | a8e6282aa418a194d0777e3fc41f8a9e2ab25381a4fb65034b53ee1fee335c05 |
| Started | 2026-10-18T12:06:36.899093Z |
| Finished | 2026-10-18T12:07:25.503995Z |
| Execution status | completed |
Every measured repetition and outcome
CPU performance
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| cpu not repeated in network supplement · repetition 1 | skipped | No measured score | Not measured |
Memory bandwidth
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| memory not repeated in network supplement · repetition 1 | skipped | No measured score | Not measured |
Disk performance
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| disk not repeated in network supplement · repetition 1 | skipped | No measured score | Not measured |
Network performance
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| Internet speed: Frankfurt, Germany (Clouvider) · repetition 1 | passed | Download throughput | 4,164 Mbit/s |
| Internet speed: Frankfurt, Germany (Clouvider) · repetition 1 | passed | Upload throughput | 1,195 Mbit/s |
| Internet speed: Frankfurt, Germany (Clouvider) · repetition 1 | passed | HTTP latency | 14.1 ms |
| Internet speed: Frankfurt, Germany (Clouvider) · repetition 1 | passed | HTTP jitter | 0.20 ms |
| Internet speed: New York, United States (2) (Clouvider) · repetition 1 | passed | Download throughput | 1,077 Mbit/s |
| Internet speed: New York, United States (2) (Clouvider) · repetition 1 | passed | Upload throughput | 308.5 Mbit/s |
| Internet speed: New York, United States (2) (Clouvider) · repetition 1 | passed | HTTP latency | 70.9 ms |
| Internet speed: New York, United States (2) (Clouvider) · repetition 1 | passed | HTTP jitter | 0.14 ms |
PostgreSQL performance
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| database not repeated in network supplement · repetition 1 | skipped | No measured score | Not measured |
Consistency over time
| Workload / repetition | Outcome | Metric | Value |
|---|---|---|---|
| consistency not repeated in network supplement · repetition 1 | skipped | No measured score | Not measured |