Logo

Platform Lab Annual Review 2021 - Shared screen with speaker view - Recording 1/2
David Ott, VMware
34:21
You are early in your reseach, but do you imagine needing new DBMS features to make your approach work well?
jeff mogul/Google
35:25
How would you construct an application that needs to be able to tolerate failures at the scale of an entire datacenter? Doesn’t this force the application to re-implement all of the Bad State Management for itself?
roland
37:31
Have you experimented with applying the principles to DBOS itself (since it’s aiming to be distributed)? A bit recursive, but might be helpful in understanding the specifics that such a DB might need, knowing that its aim is to be a holder of state.
Qian Li, Stanford
43:51
Thank you all for the questions and comments :)
Christos Kozyrakis
01:38:35
i’d like to ask a question too but I don’t know how to raise my hand :)
Francisco Romero - Stanford
01:39:08
Reactions -> Raise Hand
Israel Cidon - VMware
01:39:09
In the reaction, bottom
Manoj Wadekar
01:40:34
A question: Have you considered any use cases for distributed applications (e.g. scale-out storage) with larger messages being exchanged with RPC (e.g. Thrift)?