Self-tuning in Distributed Transactional Memory

Abstract

Many different mechanisms have been developed to implement Distributed Transactional Memory (DTM). Unfortunately, there is no "one-size-fits-all" design that offers the desirable performance across all possible workloads and scales. In fact, the performance of these mechanisms is affected by a number of intertwined factors that make it hard, or even impossible, to statically configure a DTM platform for optimal performance. These observations have motivated the emergence of self-tuning schemes for automatically adapting the algorithms and parameters used by the main building blocks of DTM systems. This chapter surveys existing research in the area of autonomic DTM design, with a focus on the approaches aimed at answering the following two fundamental questions: how many resources (number of nodes, etc.) should a DTM platform be provisioned with, and which protocols should be used to ensure data consistency.

Hybrid Transactional Replication: State-Machine and Deferred-Update Replication Combined

“…The state of any p i changes only if the value of LC , Log or any t-object changes (k b ). This can happen only when p i delivers a message through TOB, i.e., either when p i processes a transaction descriptor of a DU transaction (k a · k b , lines [56][57][58][59] or when p i processes a request which then p i executes as an SM transaction (k a · k b , lines 81-84). p i can process only one message at a time (these messages are processed as non-preemptable events).…”

“…A survey of self-tuning schemes for the algorithms and parameters used in various DTM systems can be found in [57]. A few ML-based mechanisms have been used in some of the transactional systems we discussed before.…”