On Sat, Oct 24, 2009 at 8:36 PM, MarkSwanson <mark.swanson...@gmail.com> wrote: > > I'm confused about the slide on barging: > > "txnB has a status of RUNNING and can be changed to KILLED". > > Are you implying that simply having a status of RUNNING is all that is > required for the txn to be killed? > Or, are there other requirements to "can be changed"?
There are definitely other requirements. See the bullet on that slide that says "Three conditions must be met". A big one is that txn A must have started before txn B. Older transactions are favored. > I'm having a hard time wrapping my head around the idea of the STM > just killing a fn. > Unless... the fn will be retried later. I couldn't find a good > description of what 'killing a fn' means. Ah ... I should have been more clear about that. See slide 37 that discusses the various transaction status values. I suppose you could say that KILLED is similar to RETRY. The difference is that KILLED means its going to retry because another txn barged it. > Thanks for sharing the slides. Sure! Feel free to send me more questions. -- R. Mark Volkmann Object Computing, Inc. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~----------~----~----~----~------~----~------~--~---
