At this link a table is given for detail of Zen Interactive Tuning and at this link and on ArchWiki it said:

Zen Interactive Tuning: Tunes the kernel for responsiveness at the cost of throughput and power usage.

More this means tuning responsiveness costs for throughput. On the other hand when we see the table given below the we see that for example scheduling latency has decreased which means that throughput may get higher too. Other changes in configs seem to be have direct relation with throughput too.
On a larger scale responsiveness has to have direct relation with throughput too.
The zen-kernel says that it has a higher throughput too.
Then why it is said that responsiveness has been reached at the cost of throughput?

CodePudding user response：

Your question is a little difficult to understand. You linked the Liquorix kernel and then you talk about the Zen kernel, and you mention the Zen kernel in your title.

I am going to assume you are comparing the Liqorix kernel with the stable Linux kernel since you are talking about information from that link, and you just have the names confused. The Liqorix kernel and zen kernel have similar goals which are to achieve a lower latency (more responsiveness), but that is not to say that throughput can also be achieved by tuning the kernel.

Higher throughput and lower latency are of course achievable but this comes at the cost of removing kernel features.

When it mentions that responsiveness is achieved at the cost of throughput, it is probably talking about the ratio of throughput to latency. An optimized kernel will probably have a bit more of both, but the zen and Liquorix kernels will have a ratio skewed towards latency.

I can try to clarify my answer if you need me to.