This paper by Simon Peyton Jones and others seems promising. ACID transactions are one of the most useful formalisms to come out of database research, and it's long past time they worked their way down the stack to "lower level" languages and applications than SQL and databases.
A similar approach from an operating system POV is being taken by the Coyotos project, the descendent of EROS. Note their smart use of split phase (prepare/action) kernel invocations.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment