News for hackers. Stuff that matters.

A comment on We saved millions in SSD costs by upgrading our filesystem. Parent comment

Re:We saved millions in SSD costs by upgrading our filesystem by mwcampbell on Tuesday November 09, 2021 @01:31PM (#29165435)
I would hope that for a high-throughput DB cluster like this, they're using instance-local storage rather than EBS. If that's the case, then they're probably already taking advantage of EC2 reserved instances to save a lot compared to the on-demand prices that we usually see.
[ Parent | on HN ]
Re:We saved millions in SSD costs by upgrading our filesystem by jhk727 on Tuesday November 09, 2021 @02:22PM (#29166186)
We are, though we started out using EBS. As you mentioned, NVMe instance storage performs much better for our workload. We work around the lack of durability through strong automation of point in time restore/swapping in of new nodes in case of hardware failures.

And yes, reservations make a massive difference economically.

[ Parent | on HN ]