Dariusz Wawrzyniak

From session causality to causal consistency

We discuss relationships between client-centric consistency models (known as session guarantees), and data-centric consistency models. The first group includes: read-your-writes guarantee, monotonic-writes guarantee, monotonic-reads guarantee and writes-follow-reads guarantee. The other group includes: atomic consistency, sequential consistency, causal consistency, processor consistency, PRAM consistency, weak consistency, release consistency, scope consistency and entry consistency. We use a consistent notation to present formal definitions of both kinds of consistency models in the context of replicated shared objects. Next, we prove a relationship between causal consistency model and client-centric consistency models. Apparently, causal consistency is similar to writes-follow-reads guarantee. We show that in fact causal consistency requires all common session guarantees, i.e. read-your-writes, monotonic-writes, monotonic-reads and writes-follow-reads to be preserved.

Session Guarantees to Achieve PRAM Consistency of Replicated Shared Objects

In this paper we discuss relationships between client-centric consistency models (known as session guarantees), and data-centric consistency models. Appropriate models are formally defined in the context of replicated shared objects using a consistent notation. We prove that PRAM consistency model may be achieved as a proper combination of different session guarantees.

Safety of a server-based version vector protocol implementing session guarantees

Session guarantees are used to manage replica consistency of a distributed system from the client perspective. This paper defines formally the guarantees, presents and proves safety of a protocol implementing session guarantees using server-based version vectors.

Version vector protocols implementing session guarantees

This paper analyses different protocols of session guarantees. Session guarantees (also known as client-centric consistency models) are one of the class of consistency models of replicated shared data, besides data-centric consistency models. The presentations comprises details of the data structures for the information maintained locally and passed to check consistency conditions, as well as the algorithms to process the information. The protocols are also discussed with respect to accuracy and data overhead.

Dependability Infrastructure for SOA Applications

This chapter describes two tools for improving dependability of SOA-based applications: ReSP (Reliable SOA Platform) and DyMST (Dynamic Management SOA Toolkit). ReSP is a set of modules to improve dependability in respect to availability and reliability, and to some extent safety. It is comprised of the mechanisms of reliable group communication, replication, recovery, and transaction processing. DyMST is a set of components for failure detection, monitoring and autonomic management, and distributed security policy enforcement. In order to show the dependability aspects of real applications and usage of these tools, two case studies from the medical healthcare domain are presented: Healthcare Integration Platform for the exchange of patients’ medical data among various healthcare units, and Medical Event, and Data Registering Platform for daily work improvement of medical staff.

Consistency requirements of distributed shared memory for Dijkstra's mutual exclusion algorithm

As is well known any algorithm correct in an asynchronous shared memory setting (physically shared memory) can be directly applied in distributed shared memory (DSM) systems provided that the latter guarantees strong consistency (atomic or sequential) of replicas. Generally however, in DSM systems, weaker consistency models (causal, processor, PRAM, etc.) are often considered to improve the performance. A weakening consistency model may however imply the incorrectness of the algorithm. So we face a consistency requirement problem, the problem of finding the weakest consistency model of DSM that is sufficient and necessary for algorithm correctness. We consider a reliable DSM environment, and present a complex consistency model comprising three elementary models: sequential consistency, coherence and PRAM consistency. This complex model is then applied to Dijsktras (1965) algorithm for mutual exclusion of n processes, one of the first solutions to a fundamental problem in both centralised and distributed operating systems. In the resulting algorithm, coherence and PRAM consistency are associated with some write operations performed at shared memory locations. As concurrent execution of write operations with weaker consistency models is more efficient when compared to the execution of strongly consistent operations, the proposed solution reduces synchronisation delay (mutual exclusion overhead) and thereby increases system throughput. The presented model is proven to be sufficient for algorithm correctness. Moreover, the algorithm is shown to be optimal in the sense that further relaxation of any write operations semantics violates progress (liveness) or safety of the algorithm.

The Impact of Service Semantics on the Consistent Recovery in SOA

This paper addresses a problem of consistent recovery of SOA processing. So far, the recovered state was considered as consistent if all events that have occurred before the failure were transparently recovered. However, providing such a strict consistency introduces a high performance overhead. Thus, we propose the semantic-based classification of services that enables to slack the notion of consistent recovered state from the viewpoint of services. We also present the extension of RESERVE rollback-recovery protocol that guarantees the proposed relaxed recovery consistency.

On time constraints of reliable broadcast protocols for ad hoc networks with the liveness property

In this paper we consider a formal model of ad hoc systems and its liveness property, defined with the use of the concept of dynamic sets. In this context we analyse reliable broadcast protocols dedicated for use in this kind of networks. In solutions proposed till now it is assumed that the minimum time of direct connectivity between any neighbouring nodes is much longer than maximum message transmission time. This assumption covers, however, dependence of the required minimum time of direct communication on some system parameters. Therefore, in this paper we show precisely how the minimum time of direct connectivity depends on the total number of hosts in a network and on the total number of messages that can be disseminated by each node concurrently.

Replication of Recovery Log -- An Approach to Enhance SOA Reliability

Along with development of SOA systems, their requirements in terms of fault-tolerance increase and become more stringent. To improve reliability of SOA-based systems and applications, a ReServE service, providing an external support of web services recovery, has been designed. In this paper we propose to enhance the resilience of ReServE by replication of log with recovery information, and address problems related to deployment of this solution.

Towards Relaxed Rollback-Recovery Consistency in SOA

Nowadays, one of the major paradigms of distributed processing is SOA. To improve the reliability of SOA-based systems, a ReServE service that ensures recovery of consistent processing state, has been proposed. ReServE introduces a high overhead during failure-free computing. Thus, in this paper we propose relaxed recovery consistency models that allow optimization of rollback-recovery in SOA. We propose their formal definitions, and discuss the conditions under which these models are provided by ReServE.