As far as I know, Riak does not provide any checksum on your data. I hope I am wrong.
Riak does a SHA1 of the bucket – key value ( http://lists.basho.com/pipermail/riak-users_lists.basho.com/2011-January/002820.html ), but this is not done for data integrity checking, only to get hash values to spread objects among the nodes. You can do checksums yourself (MD5, SHA) on your data, but how do you handle an error? Can you get data from the other replicas? Do you want to complicate client code with this? Wouldn’t it be best if Riak had checksums built-in, and would consider only data with correct checksums? Of course this can get complicated, as not only object reads can get corrupt: Other data, like indexes, can get corrupted too. Also, for small objects check-sums would take a lot of additional space. Google mentions the problem with data integrity in various papers: http://www.odbms.org/download/dean-keynote-ladis2009.pdf http://www.morganclaypool.com/doi/pdf/10.2200/S00193ED1V01Y200905CAC006 Runar blog.epigent.com _______________________________________________ riak-users mailing list riak-users@lists.basho.com http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
