Citation Context

...en if they replace the target boot block or the target kernel. To improve the completeness of the logger, ReVirt adapts techniques used in fault-tolerance for primaryguest application backup recovery =-=[Elnozahy02]-=-, such as checkpointing, logging, and roll-forward recovery. ReVirt is able to replay the complete, instruction-by-instruction execution of the virtual machine, even if that execution depends on non-d...

Citation Context

...We could address this limitation by including other events in SCAs (e.g., scheduling events and other I/O events) and by replaying them during verification. There is a large body of work in this area =-=[14, 13]-=- that we could leverage. 2.3 Alert generation Hosts generate SCAs when they detect an infection attempt by a worm. Vigilante enables hosts to use any detection engine provided it generates an SCA of a...

Citation Context

...aying, much of the prior work uses some form of checkpointing. There is a body of prior work on various checkpointing schemes for designing Checkpoint and Restart (CPR) systems to improve reliability =-=[10]-=-. On encountering a fault (transient error in hardware or a software error) the system can rollback to the previous checkpoint and continue execution from thereon. The purpose of checkpointing schemes...

Citation Context

...s) coordinate their local checkpoints, so that the collection of local checkpoints represents a consistent global recovery point. Coordinated system-wide checkpointing avoids both cascading rollbacks =-=[15]-=- and an output commit problem [16] for inter-node communication. Checkpoints are coordinated across the system in logical time to avoid a potentially costly exchange of synchronization messages. To en...

Citation Context

... approaches provide automatic fault tolerance for MPI applications, transparent for the user. Others approaches require the user to manage fault tolerance by adapting its application. As discussed in =-=[8, 21]-=-, many techniques of fault tolerance have been proposed in previous works, lying in various level of the software stacks : one solution is based on global coherent checkpoints, the MPI application bei...

Citation Context

...on to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. VEE’08, March 5–7, 2008, Seattle, Washington, USA. Copyright c© 2008 ACM 978-1-59593-796-4/08/03. . . $5.00 of a system just before a crash to be recovered without the need for frequent checkpoints[7, 4, 11]. Recent work has also used execution replay to efficiently collect and store software architectural traces[19]. A simple way to apply execution replay to a wide range of software is to implement execution replay for virtual machines[4]. Running softare in a virtual machine capable of being replayed allows a user to take advantage of execution replay without needing to modify software running inside the virtual machine. It also has the advantage of being able to use execution replay on an operating system kernel. In order to implement execution replay in a virtual machine, any non-deterministi...

Citation Context

...free checkpointing, but runs the risk of the domino effect [22]. Uncoordinated checkpointing is mostly used in loosely-coupled systems, where communication is infrequent and synchronization expensive =-=[6, 7, 26]-=-. However, it has also been used in tightly-coupled systems [27]. 2.2 Checkpoint Separation We group schemes into four classes based on how checkpoint data is separated from working data. Full Separat...

Citation Context

...re. Checkpoint/restart techniques for parallel jobs can be broadly classified into three categories: uncoordinated, coordinated, and communication-induced. (These approaches are analyzed in detail in =-=[10]-=-.) 2.1.1 Uncoordinated Checkpointing In the uncoordinated approach, the processes determine their local checkpoints independently. During restart, these processes search the set of saved checkpoints f...