A small question about the distributed CAP theorem?

recently, I have been reading the principle and practice of distributed consistency from Paxos to zookeeper. There is one thing I don"t understand about CAP on page 12:

clipboard.png

I may have a big problem with my understanding of the first paragraph in the picture:
1. First of all, the understanding of the sentence "avoid the problem of partition fault tolerance in the system":

 ("" ), ?
?   
,  ****?

 !

2. Next, it"s easy to understand: to avoid partitioning problems, put all the data on a distributed node, where I think it"s understandable;

3. But when it comes to "but at the same time, it should be noted that giving up P also means giving up the scalability of the system." This sentence puzzles

again.
, , ;
 p, , , ,   


it feels that "using p" will lead to a decrease in system scalability, rather than "giving up p" will lead to a decrease in system scalability.
there are a lot of problems in understanding here. Please give some advice to those who understand.


http://www.zhihu.com/question.

p refers to the tolerance of problems caused by the partition scheme, so it is called partition fault tolerance

MySQL Query : SELECT * FROM `codeshelper`.`v9_news` WHERE status=99 AND catid='6' ORDER BY rand() LIMIT 5
MySQL Error : Disk full (/tmp/#sql-temptable-64f5-1e9590f-156c.MAI); waiting for someone to free some space... (errno: 28 "No space left on device")
MySQL Errno : 1021
Message : Disk full (/tmp/#sql-temptable-64f5-1e9590f-156c.MAI); waiting for someone to free some space... (errno: 28 "No space left on device")
Need Help?