We used to, now it's all giant ZFS pools (ZFS on Linux, on 100% NVMe, 8Tb drives x 24 per unit as two RAIDz2 sets).
We chose that size for fsck times and resync times between machines. It was something we could recover in a day. Bron. On Sun, Dec 25, 2022, at 09:50, [email protected] wrote: > I understand fastmail uses 1 or 2 terabyte partitions, but im curious as to > why they chose this number? > > fsck check times? > filesystem performance degradation? > inode limitations? > fragmentation? > failure isolation? > > Also, is LVM recommended or frowned upon? > > > > > > *Cyrus <https://cyrus.topicbox.com/latest>* / Info / see discussions > <https://cyrus.topicbox.com/groups/info> + participants > <https://cyrus.topicbox.com/groups/info/members> + delivery options > <https://cyrus.topicbox.com/groups/info/subscription> Permalink > <https://cyrus.topicbox.com/groups/info/T860d16c25d61b629-M17fecf235608bd3bf3816827> -- Bron Gondwana, CEO, Fastmail Pty Ltd [email protected] ------------------------------------------ Cyrus: Info Permalink: https://cyrus.topicbox.com/groups/info/T860d16c25d61b629-M128db3f6267a8c75003ed70d Delivery options: https://cyrus.topicbox.com/groups/info/subscription
