Checkpointing Protocol Research Articles

Low-cost error containment and recovery for onboard guarded software upgrading and beyond

Message-driven (MDCD) error containment and recovery, a low-cost approach to mitigating the effect of software design faults in distributed embedded systems, is developed for onboard guarded software upgrading for deep-space missions. In this paper, we first describe and verify the MDCD algorithms in which we introduce the notion of confidence-driven to complement the communication-induced approach employed by a number of existing checkpointing protocols to achieve error containment and recovery efficiency. We then conduct a model-based analysis to show that the algorithms ensure low performance overhead. Finally, we discuss the advantages of the MDCD approach and its potential utility as a general-purpose, low-cost software fault tolerance technique for distributed embedded computing.

Impossibility of scalar clock-based communication-induced checkpointing protocols ensuring the RDT property

Communication-induced checkpointing protocols constitute an interesting approach to the on-line determination of checkpoint and communication patterns enjoying desirable properties such as domino-effect freedom. They do not add control messages to the computation, but instead may attach control information to computation messages. Among these protocols, scalar clock-based protocols are particularly attractive as they use a single integer as control information. An interesting property of checkpoint and communication patterns is Rollback-Dependency Trackability, which ensures that all local checkpoint dependencies are on-the-fly trackable. So, it would be nice to design scalar clock-based communication-induced checkpointing protocols providing the RDT property, a previously open question. This paper shows that the design of such protocols is impossible.

A checkpointing–recovery scheme for Time Warp parallel simulation

This paper presents a checkpointing–recovery scheme for Time Warp parallel simulation. The scheme relies on a checkpointing protocol, namely mixed state saving, embedding both sparse and incremental state saving modes, and on a state recovery procedure embedding both forward and backward recovery modes. This scheme is a generalization of many previous solutions, which can be obtained as particular instances of it by selecting appropriate values for the checkpointing protocol parameters. We also present two regulating algorithms to adaptively tune the checkpointing protocol parameters, in order to make the protocol reacting to variable rollback behavior. A synthetic benchmark in several different configurations has been used for evaluating and comparing our scheme with previous solutions. The obtained data show that our solution allows faster execution and, in addition, keeps quite low the amount of memory used for recording state information; this allows the scheme to not adversely affect performance when memory is a critical resource.

Rollback-Dependency Trackability: A Minimal Characterization and Its Protocol

Considering a checkpoint and communication pattern, the rollback-dependency trackability (RDT) property stipulates that there is no hidden dependency between local checkpoints. In other words, if there is a dependency between two checkpoints due to a noncausal sequence of messages (Z-path), then there exists a causal sequence of messages (C-path) that doubles the noncausal one and that establishes the same dependency.This paper introduces the notion of RDT-compliance. A property defined on Z-paths is RDT-compliant if the causal doubling of Z-paths having this property is sufficient to ensure RDT. Based on this notion, the paper provides examples of such properties. Moreover, these properties are visible, i.e., they can be tested on the fly. One of these properties is shown to be minimal with respect to visible and RDT-compliant properties. In other words, this property defines a minimal visible set of Z-paths that have to be doubled for the RDT property to be satisfied.Then, a family of communication-induced checkpointing protocols that ensure on-the-fly RDT properties is considered. Assuming processes take local checkpoints independently (called basic checkpoints), protocols of this family direct them to take on-the-fly additional local checkpoints (called forced checkpoints) in order that the resulting checkpoint and communication pattern satisfies the RDT property. The second contribution of this paper is a new communication-induced checkpointing protocol P. This protocol, based on a condition derived from the previous characterization, tracks a minimal set of Z-paths and breaks those not perceived as being doubled. Finally, a set of communication-induced checkpointing protocols are derived from P. Each of these derivations considers a particular weakening of the general condition used by P. It is interesting to note that some of these derivations produce communication-induced checkpointing protocols that have already been proposed in the literature.

On the No-Z-Cycle Property in Distributed Executions

Given a checkpoint and communication pattern of a distributed execution, the No Z-Cycle property (NZC) states that a dependency between a checkpoint and itself does not exist. In other words, a noncausal sequence of messages that starts after a checkpoint and terminates before that checkpoint does not exist. From an operational point of view, this property corresponds to the fact that each checkpoint belongs to at least one consistent global checkpoint. So it could be used, for example, for restarting a distributed application after the occurrence of a failure. In this paper we derive a characterization of the NZC property (previously an open problem). It identifies a subset of Z-cycles, namely core Z-cycles (CZCs), that has to be empty in order that the checkpoint and communication pattern of the execution satisfies the NZC property. Then, we present a communication-induced checkpointing protocol that prevents CZCs on-the-fly. This protocol actually removes the common causal part to any CZC. Finally we propose a taxonomy of communication-induced checkpointing protocols that ensure the NZC property.

Communication-based prevention of useless checkpoints in distributed computations

A useless checkpoint is a local checkpoint that cannot be part of a consistent global checkpoint. This paper addresses the following problem. Given a set of processes that take (basic) local checkpoints in an independent and unknown way, the problem is to design communication-induced checkpointing protocols that direct processes to take additional local (forced) checkpoints to ensure no local checkpoint is useless. The paper first proves two properties related to integer timestamps which are associated with each local checkpoint. The first property is a necessary and sufficient condition that these timestamps must satisfy for no checkpoint to be useless. The second property provides an easy timestamp-based determination of consistent global checkpoints. Then, a general communication-induced checkpointing protocol is proposed. This protocol, derived from the two previous properties, actually defines a family of timestamp-based communication-induced checkpointing protocols. It is shown that several existing checkpointing protocols for the same problem are particular instances of the general protocol. The design of this general protocol is motivated by the use of communication-induced checkpointing protocols in "consistent global checkpoint"-based distributed applications such as the detection of stable or unstable properties and the determination of distributed breakpoints.

Transparent checkpointing protocol for Posix.1b applications

This paper presents an overview of a proposed protocol in a Real Time Operating system to provide application-transparent fault tolerant services. Fault tolerance is achieved by saving consistent states (checkpoints) of the processes belonging to a real time application. Most of the existing checkpoint protocols are not suitable for real time environments due to the uncontrolled delay introduced when a checkpoint is saved. Our approach proposes the extension of the real time operating system services to save a checkpoint when the user performs a system call. This scheme allows the real time application designer to know the temporal specifications of every system call (checkpointing included). A real time application is composed of several Real Time processes, that share data by using the interprocess communication facilities provided by the operating system: message queues, shared memory, signals, etc. Because of these communications the operating system has to ensure the consistency of checkpoints. Current real time checkpointing schemes use global checkpointing: when a process saves a checkpoint all the processes in the system save one. The proposed solution is to track processes interactions (dependencies) in order to reduce the number of processes involved in a checkpoint.

Evaluations of domino-free communication-induced checkpointing protocols

We give a detailed analysis of communication-induced checkpointing protocols that are free of domino effect. We investigate the validity of a common intuition in the literature and demonstrate that there is no optimal on-line domino-free protocol in terms of the number of forced checkpoints. Formal proofs on comparing existing protocols in the literature are given.

Theoretical analysis for communication-induced checkpointing protocols with rollback-dependency trackability

Rollback-Dependency Trackability (RDT) is a property that states that all rollback dependencies between local checkpoints are on-line trackable by using a transitive dependency vector. In this paper, we address three fundamental issues in the design of communication-induced checkpointing protocols that ensure RDT. First, we prove that the following intuition commonly assumed in the literature is in fact false: If a protocol forces a checkpoint only at a stronger condition, then it must take, at most, as many forced checkpoints as a protocol based on a weaker condition. This result implies that the common approach of sharpening the checkpoint-inducing condition by piggybacking more control information on each message may not always yield a more efficient protocol. Next, we prove that there is no optimal on-line RDT protocol that takes fewer forced checkpoints than any other RDT protocol for all possible communication patterns. Finally, since comparing checkpoint-inducing conditions is not sufficient for comparing protocol performance, we present some formal techniques for comparing the performance of several existing RDT protocols.

Adaptive recovery for mobile environments

Mobile computing allows ubiquitous and continuous access to computing resources while the users travel or work at a client’s site. The flexibility introduced by mobile computing brings new challenges to the area of fault tolerance. Failures that were rare with fixed hosts become common, and host disconnection makes fault detection and message coordination difficult. This paper describes a new checkpoint protocol that is well adapted to mobile environments. The protocol uses time to indirectly coordinate the creation of new global states, avoiding all message exchanges. The protocol uses two different types of checkpoints to adapt to the current network characteristics, and to trade off performance with recovery time.

Consistent global checkpoints based on direct dependency tracking

We model the consistent global checkpoints in a distributed system as the maximum-sized antichains in the partially ordered set generated by the happened before relation. While the partially ordered set generated by the direct dependency tracking in uncoordinated checkpointing protocols does not capture the complete happened before relation among all checkpoints, we show that the missing happened before information does not affect the identification of consistent global checkpoints.

An efficient protocol for checkpointing recovery in distributed systems

The authors present an efficient synchronized checkpointing protocol that exploits the dependency relation between processes in distributed systems. In this protocol, a process takes a checkpoint when it knows that all processes on which it computationally depends took their checkpoints, hence the process need not always wait for the decision made by the checkpointing coordinator as in the conventional synchronized protocols. As a result, the checkpointing coordination time is substantially reduced and the possibility of total abort of the checkpointing coordination is reduced.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>