Memory management and rollback recovery in parallel architectures
Wu, Kun-Lung
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/23002
Description
Title
Memory management and rollback recovery in parallel architectures
Author(s)
Wu, Kun-Lung
Issue Date
1990
Doctoral Committee Chair(s)
Fuchs, W. Kent
Department of Study
Computer Science
Discipline
Computer Science
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Computer Science
Language
eng
Abstract
This thesis examines memory management and rollback recovery in parallel architectures. Three memory management strategies for rapid rollback recovery are studied in this thesis. The first is a cache-based error recovery scheme for shared memory multiprocessors. The second is a design for recoverable distributed shared virtual memory. The third is a twin-page disk technique for incremental checkpointing and rapid database transaction undo recovery.
In Chapter 2, a user-transparent checkpointing and recovery scheme using private caches is developed for recovering from processor transient faults. New implementation techniques using checkpoint identifiers and recovery stacks are examined as a means to reducing performance degradation. This multiprocessor cache-based checkpointing technique prevents rollback propagation, provides for rapid recovery and can be integrated into standard cache coherence protocols.
In Chapter 3, we examine the problem of rollback recovery in distributed shared virtual memory environments, where the shared memory is implemented by software in a loosely coupled distributed multicomputer system. A user-transparent checkpointing recovery scheme is presented for designing recoverable distributed shared virtual memory. The checkpointing scheme is unique in that it can be integrated with the memory coherence protocol for managing the shared virtual memory. The recoverable distributed shared virtual memory allows the system to restart computation from a checkpoint without a global restart.
In Chapter 4, a twin-page disk scheme is presented for incremental checkpoint and fast rollback of parallel processes in the recoverable distributed shared virtual memory developed in Chapter 3. Also the scheme is examined as a means for rapid database transaction undo recovery. The twin-page scheme allows incremental flushing of dirty pages from main memory to disk without the requirement of an explicit disk undo at the time of recovery. Undo is implicitly performed during normal processing by not fetching the invalid data.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
