Partial loading of XMI models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

XML Metadata Interchange (XMI) is an OMG-standardised model exchange format, which is natively supported by the Eclipse Modeling Framework (EMF) and the majority of the modelling and model management languages and tools. Whilst XMI is widely supported, the XMI parser provided by EMF is inefficient in some cases where models are readonly (such as input models for model query, model-to-model transformation, etc) as it always requires loading the entire model into memory. In this paper we present a novel algorithm, and a prototype implementation (SmartSAX), which is capable of partially loading models persisted in XMI. SmartSAX oérs improved performance, in terms of loading time and memory footprint, over the default EMF XMI parser. We describe the algorithm in detail, and present benchmarking results that demonstrate the substantial improvements of the prototype implementation over the XMI parser provided by EMF.

Original languageEnglish
Title of host publicationProceedings - 19th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2016
PublisherACM
Pages329-339
Number of pages11
ISBN (Electronic)9781450343213
DOIs
Publication statusPublished - 2 Oct 2016
Event19th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2016 - Saint-Malo, France
Duration: 2 Oct 20167 Oct 2016

Conference

Conference19th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2016
Country/TerritoryFrance
CitySaint-Malo
Period2/10/167/10/16

Keywords

  • EMF
  • Partial model loading
  • XMI

Cite this