CPU RESOURCE POLICY

CPU Policy Update

The existing Relaxed Mode and Strict Mode rules remain unchanged. The following two rules apply in addition:

• Additional rule 1: Two consecutive CPU performance limits within a rolling 48-hour (two-day) window result in service suspension. After suspension, the user may open a support ticket to request restoration.
• Additional rule 2: Two consecutive suspensions for CPU limits within a rolling 168-hour (seven-day) window constitute CPU abuse. All VPS instances under the account will be deleted, and the account will also be deleted.

Misconception 1: “CPU is permanently capped at 40%”

Our mechanism provides 100% CPU performance before throttling. Once a limit is triggered, CPU performance is capped at 15% in Strict Mode or 30% in Relaxed Mode.

Under the existing baseline rules, every user can run at full performance for at least 45 minutes. We consider this reasonable within the industry. Figures such as 40% on this page are observation thresholds, not a fixed CPU performance cap imposed when the service is activated.

Some providers using PVE impose an initial CPU limit equal to a percentage of one core (s%), so peak performance is restricted from the outset. Our mechanism instead makes full CPU performance available before a limit is triggered.

At purchase, all plans already state that CPU resources are fairly allocated and shared, not dedicated. Shared VPS products are not intended for continuous, sustained heavy CPU workloads.

Misconception 2: “Four occurrences and the VPS is deleted!”

The phrase “deleted after four occurrences” leaves out a crucial condition: within a rolling 168-hour (seven-day) window. This is not a lifetime count of four throttling events. It concerns repeated limits that result in two suspensions within the specified rolling window.

Normally, when a suspended user opens a ticket to request restoration, we proactively help with a brief investigation. The purpose is to check that malicious scripts are not consuming CPU without the user’s knowledge. Some scripts include additional background tasks that may continuously use CPU resources.

At a minimum of 45 minutes of full performance across all allocated cores for each occurrence, two suspensions within a rolling 168-hour (seven-day) window represent at least 180 minutes of full-core CPU use. If this continues to occur, we consider the two main explanations to be:

1. A substantial CPU requirement: The workload is unsuitable for our shared VPS products and should be moved to a dedicated server.
2. Repeated exploitation by malicious scripts: Background tasks are consuming CPU, so scripts and running processes need to be checked.

We regard these as the only two explanations; there is no third possibility. Reducing the policy to “deleted after four occurrences” while omitting the seven-day rolling window and the two-suspension condition takes the policy out of context and does not accurately describe its operation.

Observed Frequency of Enforcement

When the rule explanation was published on September 12, 2026, fewer than 10 instances out of several thousand had triggered this restriction. On that basis, we considered the limits sufficiently generous to avoid affecting normal lightweight use. These figures describe the observations at that publication date.