Re: Is this model correct?

[Rick James]

No problem. The space for the index(es) can be larger than for the data.

With InnoDB, the PRIMARY KEY is incorporated into the data, so it registers as very little space. Any secondary keys (you have one) include all the fields of the PRIMARY KEY. So,
KEY "id_indexkey" ("ck_key")
actually contains all the fields of your table. (But it is sorted first on ck_key.)

This implies that the Index_length would be about the same as the Data_length, which it is. Since both data and index are stored in BTrees, there is some variation due to packing and block splits. Hence, one could happen to be bigger or smaller than the other. Yours differ by 2%. I would not be alarmed at a 30% difference (based purely on how BTrees work).