I had a problem with growing Domino databases and didn’t have the option to turn DAOS. And I thought how could this problem be solved if I did some database design. What is the problem today with large Domino databases or large attachments today?
1. Large or many Attachments makes the database slow
2. Opening documents become slow.
DAOS solves these issues by extracting the attachment out of the database and creating a file with a hash name to get a single copy solution, Very clever. The hashing of files solution original developed by and MIT professor named Ron Rivest. He was one of the founders of RSA (the R is from his name )
I tough if you could have a DAOS like solution stored inside a single NSF file then you could benefit from this when if you server didn’t have DAOS enabled. And also if you did a local replica.
If the attachments were extracted from the actual document like they are in DAOS, I belive that the performance would increase. This because the attachments are only touched in they are needed and not every time the document fetched like today. But this is just a wish from me 😉
It will be interesting to see what new features IBM will implement into the NSF container in the next ODS version. So the Domino NSF container can show that it’s the largest NoSQL database what I know of.
Have a great NSF Day!!