Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Presentation by Michał Dykas on how to optimize SSR performance

Michał Dykas and PCG team shared challenges and solutions on reducing lag times when dealing with large repositories in DSpace.

  • PCG team ran into some scalability/performance issues with a large repository.  Currently documented Performance Tuning DSpace instructions were not helpful enough for a large site
  • Developed a way to use Redis for caching (instead of built-in DSpace caching which uses an in-memory lru-cache).  This Redis solution worked well for a large site
  • PCG will donate Redis integration code & also make suggestions on improving documentation.  (This Redis approach would be optional, only for sites that need it.  However for larger sites it can help alleviate the need for a large cluster of node instances