The British Library holds over 13 million books, 920,000 journal and newspaper titles, 57 million patents, 3 million sound recordings, and much more.
The British Library was developing a single entry point for a number of its web resources. The single entry point was a turning point for the Library’s website, enabling users to search much more of the Library’s diverse content from its home page at www.bl.uk.
The new approach to searching was anticipated as having a considerable impact on the four web resources being searched – the Library’s main website; its catalogue; the Collect Britain site; and its database of journal articles. These four resources are all very different websites with different users and utilising different technical platforms and search engines. Each one also had users who were not routinely searching the other resources. With the introduction of the single entry point search, all the resources would be accessible to all searchers.
The Library required the individual; resources and the whole system to be load tested prior to launch of the single entry point. Weak points in the system needed to be identified along with the levels of usage at which they fail.
Acutest were selected after a competitive tender. The end-to-end load testing solution included:
- Benchmarking each of the web resources searched by the entry point
- Optimising the sustainable throughput/saturation point of the resources to identify bottlenecks and assist with tuning
- Identifying and optimising the saturation point and latency/throughput profile of each resource
- Characterising the behaviour of each catalogue under selected usage scenarios
- Testing the behaviour of the system in overload conditions where the levels of demand being applied exceeded the saturation point and sustainable throughput point.
Acutest identified a number of bottlenecks and weak points and were proactive in assisting the Library with the required fixes and the retesting. The British Library’s single entry point launched successfully.