Hacker Newsnew | past | comments | ask | show | jobs | submit | neal's commentslogin

Good point!

I'm not sure if there are any databases that do your 'with a lot of care' option, but for anyone curious about what that might look like in practice there are file systems that forgo write-ahead logging and maintain metadata consistency using techniques like soft updates[0] or copy-on-write up to a root[1].

[0]: https://www.usenix.org/conference/1999-usenix-annual-technic... [1]: https://www.cs.hmc.edu/~rhodes/courses/cs134/fa20/readings/T... (yes, ZFS can be configured to use a WAL too for durability)


> I'm not sure if there are any databases that do your 'with a lot of care' option

LMDB springs to mind.

> LMDB was designed to resist data loss in the face of system and application crashes. Its copy-on-write approach never overwrites currently-in-use data. Avoiding overwrites means the structure on disk/storage is always valid, so application or system crashes can never leave the database in a corrupted state.

https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Databa...


> yes, ZFS can be configured to use a WAL too for durability

ZFS always uses ZIL for sync writes. You can optionally set `sync=disabled` for a dataset to switch it off. I would not describe it as "can be configured to use WAL".


Graduating in September with a MS in computer science.

I'm looking for a software development position. My research area was computer networks, but interested in other areas as well. More interested in working on back-end type of stuff than front-end/UI. In Boston area, but willing to relocate.

http://nealc.com


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: