incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Cipar <jci...@cmu.edu>
Subject Re: Consistency model
Date Sun, 17 Apr 2011 17:03:39 GMT
> For a second, I thought this thread was saying I could see value(s) < new value(s)
within the same process on the second read at quorum

That's exactly what I'm saying.  Within a single process I see this behavior, when reading
with consistency_level=QUORUM

Read value 1
Read value 2
Read value 1  # uh oh!  we've gone backwards



On Apr 17, 2011, at 12:15 PM, William Oberman wrote:

> Cool, that is exactly what I was thinking/expecting, and I feel better about the assumptions
I've used in my refactor to move from MySQL -> cassandra.  
> 
> I basically gave up on trying to force my code to assume (even at quorum): 
> -read value(s) for key from cassandra
> -update/add/delete columns for key in memory based on value(s)
> -write new value(s) for key to cassandra
> That the next read for key in the same process would be equal to the new value(s).  But
I do assume I read value(s) >= the new value(s). Of course, in another process, I don't
assume anything (e.g. my application is ok reading the same state A in two processes and updating
the state to either B or C.  I figured that's the best a distributed algorithm could do without
a "test and set" operation).
> 
> For a second, I thought this thread was saying I could see value(s) < new value(s)
within the same process on the second read at quorum, which would have been bad for me.  But
this thread is not saying that, so I'm still happy with cassandra :-) 
> 
> will
> 
> On Sun, Apr 17, 2011 at 11:24 AM, Milind Parikh <milindparikh@gmail.com> wrote:
> William 
> The issue is regarding whether you will see A or B; with any guarantee of either. The
discussion implies no; until the QW is complete.
> 
> /***********************
> sent from my android...please pardon occasional typos as I respond @ the speed of thought
> ************************/
> 
> 
>> On Apr 17, 2011 7:48 AM, "William Oberman" <oberman@civicscience.com> wrote:
>> 
>> I'm pretty new to all of this, and I'm in the process of building my mental model
of Cassandra, but I'm still feeling better about this thread. The way I figure it:
>> 1. I'm trying to mutate the state of a key's column from A to B from a thread somewhere
(quorum)
>> 2. I'm trying to read the state of a key from a thread somewhere else (quorum)
>> 
>> If #1 succeeds I'm guaranteed to see B. If #1 fails (with an exception) I'll see
either A or B. I think I was concerned about that, and wanted to see A in #2 until success
in #1.  But, I wanted to get to state B, and if #1 retries until guaranteed success, do I
care if I set B earlier than I expected?  I'm thinking no. 
>> 
>> I guess in terms of distributed algorithms/reasoning about systems, I'm feeling ok
with this level of guarantee (again, given the failed write tells the client code of the undefined
state). 
>> 
>> 
>> 
>> On Apr 17, 2011, at 10:10 AM, James Cipar <jcipar@cmu.edu> wrote:
>> 
>> > I'm pretty new to Cassandra, ...
>> 
> 
> 
> 
> 
> -- 
> Will Oberman
> Civic Science, Inc.
> 3030 Penn Avenue., First Floor
> Pittsburgh, PA 15201
> (M) 412-480-7835
> (E) oberman@civicscience.com


Mime
View raw message