与ZooKeeper的Cassandra交易 - 这有用吗？

I am trying to implement a transaction system for Cassandra with the help of ZooKeeper. Since I don't think I have enough experience in database implementation, I would like to know if my idea would work in principle, or is there any major flaw.

我试图在ZooKeeper的帮助下为Cassandra实现一个事务系统。由于我认为我没有足够的数据库实施经验，我想知道我的想法是否原则上有效，或者是否有任何重大缺陷。

Here is the high level description of the steps:

以下是步骤的高级描述：

1. identify all the rows(keys) and columns to be edited. Let the keys be [K0..Kn]
2. 识别要编辑的所有行（键）和列。让键为[K0..Kn]
3. apply write lock on all the rows involved (locks are in-memory Zookeeper implementation)
4. 对所有涉及的行应用写锁定（锁是内存中的Zookeeper实现）
5. copy the old values to separate locations in Cassandra which are uniquely identified by key: [K'0..K'n]
6. 将旧值复制到Cassandra中由键唯一标识的不同位置：[K'0..K'n]
7. store [K'0..K'n] and the mapping of them to [K0..Kn] in ZooKeeper using persistent mode
8. 使用持久模式存储[K'0..K'n]并将它们映射到ZooKeeper中的[K0..Kn]
9. go ahead apply the update to the data
10. 继续将更新应用于数据
11. delete the entries in ZooKeeper
12. 删除ZooKeeper中的条目
13. unlock the rows
14. 解锁行
15. delete the entries of [K'0..K'n] lazily on a maintenance thread (cassandra deletion uses timestamp, so K'0..K'n can be reused for another transaction with a newer time stamp)
16. 在维护线程上懒洋洋地删除[K'0..K'n]的条目（cassandra删除使用时间戳，因此K'0..K'n可以重新用于具有较新时间戳的另一个事务）

Justification:

理由：

1. if the transaction failed on step 1-4, no change is applied, I can abort the transaction and delete whatever is stored in zookeeper and backup-ed in cassandra, if any.
2. 如果在步骤1-4中事务失败，则不应用任何更改，我可以中止事务并删除存储在zookeeper中的任何内容并在cassandra中备份，如果有的话。
3. if the transaction failed on step 5, the information saved on step 3 is used to rollback the any changes.
4. 如果在步骤5中事务失败，则在步骤3中保存的信息用于回滚任何更改。
5. if the server happen to be failed/crashed/stolen by cleaning man, upon restart before serving any request, I check if there is any keys persisted in the zookeeper from step 4, if so, i will use those keys to fetch backed up data stored by step 3, and put those data to where they were, thus roll-back any failed transactions.
6. 如果服务器碰巧被清理人员发生故障/崩溃/被盗，在重新启动服务器任务请求之前，我会检查动物园管理员是否在步骤4中保留了任何密钥，如果是，我将使用这些密钥来获取备份数据由步骤3存储，并将这些数据放到它们所在的位置，从而回滚任何失败的事务。

One of my concern is what would happen if some of the servers are partitioned from the cluster. I have no experience in this area, does my scheme work at all? and does it work if partition happens?

我担心的一个问题是，如果某些服务器是从群集中分区的，会发生什么。我没有这方面的经验，我的计划是否有效？分区发生时是否有效？

1 个解决方案

#1

You should look into Cages: http://ria101.wordpress.com/2010/05/12/locking-and-transactions-over-cassandra-using-cages/

您应该查看Cages：http：//ria101.wordpress.com/2010/05/12/locking-and-transactions-over-cassandra-using-cages/

http://code.google.com/p/cages/

http://code.google.com/p/cages/