Hi John,

I had exactly the same reflexions.

I'm using zookeeper and cage to lock et isolate.

but how to rollback? 
It's impossible so try replay!

the idea is explained in this presentation http://www.slideshare.net/mattdennis/cassandra-data-modeling (starting from slide 24)

- insert your whole data into one column
- make the job
- remove (or expire) your column.

if there is a problem during "making the job", you keep the possibility to replay and replay and replay (synchronously or in a batch).



2011/12/5 John Laban <john@pagerduty.com>

I'm building a system using Cassandra as a datastore and I have a few places where I am need of transactions.

I'm using ZooKeeper to provide locking when I'm in need of some concurrency control or isolation, so that solves that half of the puzzle.

What I need now is to sometimes be able to get atomicity across multiple writes by simulating the "begin/rollback/commit" abilities of a relational DB.  In other words, there are places where I need to perform multiple updates/inserts, and if I fail partway through, I would ideally be able to rollback the partially-applied updates.

Now, I *know* this isn't possible with Cassandra.  What I'm looking for are all the best practices, or at least tips and tricks, so that I can get around this limitation in Cassandra and still maintain a consistent datastore.  (I am using quorum reads/writes so that eventual consistency doesn't kick my ass here as well.)

Below are some ideas I've been able to dig up.  Please let me know if any of them don't make sense, or if there are better approaches:

1) Updates to a row in a column family are atomic.  So try to model your data so that you would only ever need to update a single row in a single CF at once.  Essentially, you model your data around transactions.  This is tricky but can certainly be done in some situations.

2) If you are only dealing with multiple row *inserts* (and not updates), have one of the rows act as a 'commit' by essentially validating the presence of the other rows.  For example, say you were performing an operation where you wanted to create an Account row and 5 User rows all at once (this is an unlikely example, but bear with me).  You could insert 5 rows into the Users CF, and then the 1 row into the Accounts CF, which acts as the commit.  If something went wrong before the Account could be created, any Users that had been created so far would be orphaned and unusable, as your business logic can ensure that they can't exist without an Account.  You could also have an offline cleanup process that swept away orphans.

3) Try to model your updates as idempotent column inserts instead.  How do you model updates as inserts?  Instead of munging the value directly, you could insert a column containing the operation you want to perform (like "+5").  It would work kind of like the Consistent Vote Counting implementation: ( https://gist.github.com/416666 ).  How do you make the inserts idempotent?  Make sure the column names correspond to a request ID or some other identifier that would be identical across re-drives of a given (perhaps originally failed) request.  This could leave your datastore in a temporarily inconsistent state, but would eventually become consistent after a successful re-drive of the original request.

4) You could take an approach like Dominic Williams proposed with Cages:  http://ria101.wordpress.com/2010/05/12/locking-and-transactions-over-cassandra-using-cages/    The gist is that you snapshot all the original values that you're about to munge somewhere else (in his case, ZooKeeper), make your updates, and then delete the snapshot (and that delete needs to be atomic).  If the snapshot data was never deleted, then subsequent accessors (even readers) of the data rows need to do the rollback of the previous transaction themselves before they can read/write this data.  They do the rollback by just overwriting the current values with what is in the snapshot.  It offloads the work of the rollback to the next worker that accesses the data.  This approach probably needs an generic/high-level programming layer to handle all of the details and complexity, and it doesn't seem like it was ever added to Cages.

Are there other approaches or best practices that I missed?  I would be very interested in hearing any opinions from those who have tackled these problems before.