ZOOKEEPER-2349:Update documentation for snapCount

[maoling]

Origin patch can not apply now and it has something inaccurate.so I give it a new PR.
more details in JIRA

[afine]

I think this is an excellent documentation improvement, but we still may be able to phrase this in a clearer way. It may be good to reiterate that ZooKeeper writes to a transaction log between snapshotting here, so the reader understands the purpose of the snapCount.

Perhaps something like:

ZooKeeper records its transactions using snapshots and a transaction log (think write-ahead log). The number of transactions recorded in the transaction log before a snapshot can be taken (and the transaction log rolled) is determined by snapCount. In order to prevent all of the machines in the quorum from taking a snapshot at the same time, each ZooKeeper server will take a snapshot when the number of transactions in the transaction log reaches a runtime generated random value in the [snapCount/2, snapCount) range.

[maoling]

• awesome suggestions. especially emphasizing the purpose of this snapshot strategy is to prevent all of the machines in the quorum from taking a snapshot at the same time
• done

[hanm]

Committed to master 563f892,
branch-3.5 4051b38,
branch-3.4 574a94d