Low overhead communication‐induced checkpointing protocols ensuring rollback‐dependency trackability property

Abstract

SummaryCommunication‐induced checkpointing (CIC) protocols aim at finding consistent global checkpoint from which a system can safely restart without the risk of domino effect. CIC class gives processes the maximum autonomy for taking their local checkpoints while directs them to take forced checkpoints when harm patterns are detected. Such patterns can constitute of non‐causally doubled Z‐paths that violate the causal relation and can cause checkpoints uselessness, which results in hidden dependencies between checkpoints. Ensuring the absence of hidden dependencies hence means the ensure of the Rollback‐Dependency Trackability (RDT) property.Reducing the control information overhead is the main aim of our proposals. Our first proposal is called CSFDAS (Constant Size Fixed Dependency After Send), and the second is called CSRDTParner (Constant Size RDTPartner). Unlike to the previous works, the new proposed protocols use direct dependency clocks instead of transitive dependency vectors while keeping only a constant size of control information on messages. Conducted simulations show that the new protocols achieve a good performance compared to the RDT protocols proposed so far in the literature.

Scalable Communication-Induced Checkpointing Protocol with Little Overhead for Distributed Computing Environments

“…Most existing CIC protocols [3][4][5][6][7][8][9][10]14,15] have been designed to ensure that no basic checkpoint becomes useless by satisfying Theorem 2 at all times.…”

“…In addition, if needed before message delivery, additional checkpoints, called forced checkpoints, can be taken based on the local variables of the process and the information piggybacked on the messages [2]. Most CIC protocols [3][4][5][6][7][8][9][10][11][12][13][14][15] with these advantages have the following common behavioral features. If each process recognizes the delivery of a message may make any other local checkpoint of another process, called a basic checkpoint, become useless, the protocols make the process take a forced checkpoint.…”

“…CIC protocols are classified into two types, model-based CIC and timestamp-based CIC [1]. Although model-based protocols [3,4] may be transformed to their corresponding timestamp-based ones [5][6][7][8][9][10][11][12][13][14][15], the overhead of the first, in its eagerness to break suspicious message exchange patterns, can be considerably higher than that of the second in terms of the number of forced checkpoints [1]. Timestamp-based protocols have continuously advanced with their protocol-specific checkpoint timestamping schemes using Lamport's logical clock to decrease the number of forced checkpoints in the following manner.…”

“…Although developed to operate based on a reliable first-in, first-out (hereafter, FIFO) communication channel, most traditional CIC protocols [3][4][5][6][7][8][10][11][12][13][14][15] do not use this advantageous feature to lessen this sort of forced checkpointing overhead. A previous CIC protocol [9] attempted to decrease the number of forced checkpoints by obtaining the up-to-date local clock of the next checkpoint of every other process based on performing sender-based message logging earlier than before.…”

“…The protocol can overcome the constraint of the piecewise deterministic model by recognizing which execution point is deterministic in each checkpoint interval. However, it has the same limitation as the traditional ones [3][4][5][6][7][8][11][12][13][14][15] mentioned above-that is, the slow gathering of the most recent local timestamp of the next checkpoint of every other process. In addition, if non-deterministic events that cannot be logged often occur in the execution of processes, the performance gain may be greatly diminished.…”