Published November 26, 2019 | Version v1

Comparing the Performance of OAI-PMH with ResourceSync

  • 1. Los Alamos National Laboratory, USA;
  • 2. Los Alamos National Laboratory, USA; Knowledge Media institute, The Open University, United Kingdom;
  • 3. Los Alamos National Laboratory, USA; Knowledge Media institute, The Open University, United Kingdom; Knowledge Media institute, The Open University, United Kingdom

Description

In this work, we conduct experiments comparing OAI-PMH with ResourceSync along a set of dimensions, scenarios and implementation setups. Our results are the first to disclose that OAI-PMH performance varies significantly across platforms, explaining why it is difficult for aggregators to stay synchronised with data providers. We also provide evidence showing that ResourceSync is, if applied in suitable synchronisation mode, substantially faster and more effective in synchronisation of resources and puts also lower load on the data providers than OAI-PMH. By comparing the performance of OAI-PMH with ResourceSync on the task of metadata harvesting (small files), i.e. the most challenging scenario for ResourceSync, we demonstrate that ResourceSync performs up to about 10x faster in the Batch mode than an average OAI-PMH repository.

Files

submission abstract.txt

Files (896 Bytes)

Name Size Download all
md5:53083c175ce844c6de0500377293f18c
896 Bytes Preview Download