Proficient in NoSQL come in and see what's the best solution for this situation? What to do with it

Proficient in NoSQL come in and see what's the best solution for this situation??
Site A's database has a very simple table xx, field:
ID: Self-increment
Key: The value is the only character
Value: values are arbitrary strings

Website B needs to read the data of XX table in this website a every day.


What I'm doing now is:

Whenever site a operates the XX table, it synchronizes the same operation to the MySQL database in remote server C. This way, site B reads data from remote server C every day for query display. However, the performance of this method is still a problem, the query needs to connect to the remote server C MySQL database, although the site B after the query to get the data in the local cache, but I do not feel satisfied.


Recently found that NoSQL should be more suitable for storing my xx table, so if you use NoSQL words should be how to do better?


I also thought of a method:

When site a operates XX table, each insert new record or modify record, will be in a directory specifically for each record generated a corresponding TXT file, the filename is key, the file content is value. In this way, every time the site B query data, curl site A directory has no corresponding TXT file name, if any, the value is displayed. The disadvantage of this approach is that a very large number of small files are generated under a directory of site A.

Or, is there any good way to save the data in the XX table to a file? And then pass it to Site B? Then Site B operates on a local query. The disadvantage is that every time the site XX table updates and other operations, you need to sync this file to site B over there.

------Solution--------------------
Memcache is not a good solution to the problem ... or that redis.
Several servers share data the best way. But the data can not be huge, the boss said the memory is very expensive. Ha ha

