Open Transactions on Shared Memory

Abstract

Transactional memory has arisen as a good way for solving many of the issues of lock-based programming. However, most implementations admit isolated transactions only, which are not adequate when we have to coordinate communicating processes. To this end, in this paper we present OCTM, an Haskell-like language with open transactions over shared transactional memory: processes can join transactions at runtime just by accessing to shared variables. Thus a transaction can co-operate with the environment through shared variables, but if it is rolled-back, also all its effects on the environment are retracted. For proving the expressive power of TCCS we give an implementation of TCCS, a CCS-like calculus with open transactions

Software Transactional Memory with Interactions

“…This work builds on the calculus with shared memory and open transactions described in [14]. In loc.…”

A Specification of Open Transactional Memory for Haskell

“…This work builds on the ideas in [10] where we described an abstract calculus with shared memory and open transactions. In loc.…”

BiGGer: A Model Transformation Tool written in Java for Bigraph Rewriting in GrGen.NET

“…Specifically, one substructure of a bigraph is a forest, which is relevant to the problem of locating forests in trees (Bacci et al, 2014), while the other substructure is a hypergraph, which is associated with the subhypergraph matching problem. To date, practical bigraph matching primarily exists as a CSP implementation (Miculan & Peressotti, 2014), or SAT-based algorithm (Sevegnani et al, 2010) in the literature.…”

“…There are few practical solutions concerned with the bigraph matching problem for various kinds of bigraphs; a non-exhaustive presentation is the following. For binding bigraphs (i.e., links have local scopes) by an inductive characterization of matching (Birkedal et al, 2007;Damgaard et al, 2013); for directed bigraphs (which subsume pure bigraphs) (Bacci et al, 2009); for bigraphs with sharing (i.e., the place graph is a directed acyclic graph) by using a SAT-based algorithm (Sevegnani & Calder, 2016); for the pure case by (Miculan & Peressotti, 2014) as a CSP, and further an adapted reduction of the problem for directed bigraphs to a CSP (Chiapperini et al, 2020); and the work in Gassara et al (2019), which proposes a toolchain for bigraph matching that is conceptually most similar to our approach but not actively developed anymore and not as efficient as GrGen.NET.…”