Wednesday, 11 October, 2023 - 11:50–12:30
Danny Kopping, Grafana Labs
Abstract:
Our cloud database stores billions of files in object storage. We were encountering severe rate-limits when attempting to run large queries, and we needed a way to cache a large subset in a cost-effective manner without hurting performance.
In this talk, I'll describe how we're using memcached in combination with local SSDs provided by major cloud vendors to scale up our memcached to tens of terabytes, improving reliability, and saving money!
BibTeX
@conference {292080,
author = {Danny Kopping},
title = {Cache Me If You Can: How Grafana Labs Scaled Up Their Memcached 42x \& Improved Reliability},
year = {2023},
address = {Dublin},
publisher = {USENIX Association},
month = oct
}
author = {Danny Kopping},
title = {Cache Me If You Can: How Grafana Labs Scaled Up Their Memcached 42x \& Improved Reliability},
year = {2023},
address = {Dublin},
publisher = {USENIX Association},
month = oct
}
