D. Hulse, "Store Architecture in a Persistent Operating System", Ph.D., Department of Computer Science, University of Adelaide, Adelaide, 1998. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Concurrency - The fly in the ointment? - Blackburn, Zigman (1999) (2 citations) (Correct)
....causal dependency checkpoint scheme that does not require the synchronization of all interdependent computations. However, this approach depends on the concurrent processes taking their checkpoints at times that can provide consistent cuts it will only find consistent cuts, not enforce them [Hulse 1998]. It is therefore necessary for the programmer or some other agent, such as an operating system daemon, to ensure that checkpoints are taken in such a fashion as to guarantee consistent cuts and therefore guarantee recoverability. Both approaches are implemented within the operating system kernel ....
HULSE, D. 1998. Store Architecture in a Persistent Operating System. PhD thesis, University of Adelaide.
Lumberjack: A Log-Structured Persistent Object Store - Hulse, Dearle, Howells (1998) (1 citation) Self-citation (Hulse) (Correct)
....operation. The result is that almost 100 of the time taken to perform a snapshot can be attributed to writing modified object data to the log. In contrast, the cost of snapshot operations in stores based on shadow paging can easily be dominated by the overhead of writing back the meta data [6]. Many systems [5, 8, 15] technically known as log based systems, store objects in a fixed location on disk and use a log to increase the efficiency of snapshot operations by writing data sequentially rather than randomly as would otherwise be required. The data written to the log is copied to ....
D. Hulse, "Store Architecture in a Persistent Operating System", Ph.D., Department of Computer Science, University of Adelaide, Adelaide, 1998.
Persistent Operating System Support for Persistent.. - O'Lenskie, Dearle, Hulse (1999) Self-citation (Hulse) (Correct)
....CORBA objects after every update and before the update has propagated to CORBA clients. This approach may be sophisticated enough for many applications since snapshot costs are low, especially when performed frequently due to Grasshopper s log structured storage technology employed by the kernel [5]. However, it is also possible to log incoming requests and use these to restore the system to an externally visible state during recovery. This approach depends on the ability to restore a computation to some snapshotted state. The recovery manager can pull the state of the container forward in ....
D. Hulse, "Store Architecture in a Persistent Operating System", , Computer Science Department, University of Adelaide, Adelaide, 1998. URL ftp://persistence.cs.stir.ac.uk/pub/papers/davethesis. pdf
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC