Collect common system metrics like load average, CPU load in userspace and kernel, IO wait, etc... inside ClickHouse.

Use case
Many users don't have any monitoring tools set up. Some cloud providers have basic monitoring capabilities by default but they are often not enough. Collecting and storing system metrics inside ClickHouse will be essentially free.

Another imaginary use case - is to run empty ClickHouse server just to collect system metrics.

Describe the solution you'd like
Calculate the required values in AsynchronousMetrics.

There are multiple tools and libraries that have relevant code: NetData, PCP.

Unfortunately, most of them are under GPL or LGPL. We can use LGPL inside ClickHouse if the code will be represented as a separate library (not as just a chunks of reused code).