Rob Coli <rcoli <at> digg.com> writes: > Is your workload straight INSERT or does it contain UPDATE and/or > DELETE? If your workload contains UPDATE/DELETE and GCGraceSeconds (10 > days by default) hasn't passed, you might have a non-trivial number of > tombstone rows. Only major compactions clean up tombstones in the > current implementation and even given improvements [1] they will always > be necessary in some cases. > > =Rob
I am in the testing phase so I'm writing dummy data using keys numbered '1' to '1000000', writing each row exactly once. I'm using batch_mutate() but in this particular case my batch size is 1. I do not do any deletes and I am not overwriting any rows since I write each key (1-1,000,000) only once. So I don't think I should have any tombstones, unless there's something going on that I don't know about. Thanks for your help, Julie
