1a1a11a

It is such as shame: we tell you that we are writing a paper on this exact topic and you created a GitHub and advertises everywhere

"Tectonic (Pan et al., 2021), Meta’s general-purpose distributed file system, is used to build a storage fabric (Battey and Gupta, 2024) for Llama 3 pre-training. It offers 240 PB of storage out of 7,500 servers equipped with SSDs, and supports a sustainable throughput of 2 TB/s and a peak throughput of 7 TB/s"

I would expect at least 10s if not 100s TB/s from a cluster of 7500 servers with SSDs, what is the bottleneck?

Got it. Make sense, I like the reason why you want to make it circular. :)

it looks like the ring is implemented using a linked list?

Great point! We tried the oracle (Belady MIN) and found that the oracle often does not admit new objects (<20% chance admitting new objects) in most traces. That's why we need to evict new objects quickly

This is an awesome reply that I would like to see here. Do you mind sharing more details on what type of cache or use case needs the performance you mentioned? Database bufferpool?

> a lot of these practical details have never been written down a far as I can tell, it is passed along as lore and beer convos

I wish the valuable knowledge could be written down and shared.

Yup, I am on the same page as you, after seeing the many state-of-the-art having similar miss ratios, I am inclined to think we are close to the maximum possible hit ratio bounded by the limited information from the request history (information theory).

I love your comments!

This is fantastic work!