Does this actually work?

[dpalex]

Hi, I was wondering if PrimoCache actually works? It seems to be extremely slow. I have a 1TB Intel 600p I use for L2 cache and reading a file using Parkdale shows roughly 250-270MBps, shouldn't this be a lot higher considering my 600p is capable of 1800MBps sequential read?

Here's a screencast showing me loading the file using Parkdale and showing that the L2Storage Read is being hit.
https://screencast-o-matic.com/watch/cbi105loMk

Any ideas?

[Support]

What's the PrimoCache version? If it is not 3.0b, could you try the 3.0beta and see if there's difference? Thank you!
viewtopic.php?f=32&t=4395

[dpalex]

Hi, I am using 3.0.0 beta.

I specifically picked a benchmark which does not rely on creating temporary files and reading them. And according to the screencast I made, the cache is definitely hit.

[Support]

1.

considering my 600p is capable of 1800MBps sequential read

Does this result comes from the same benchmark tool under the same test condition?

2. I see the cache hit rate is 63.3%, I'm not sure if you have read other data before the testing. Could you reset the statistics before the test and then do the test? I'd like to know the speed on 100% cache hit.

[dpalex]

I managed to run another benchmark which showed better results. The file benchmark of HD Tune Pro. I suppose I see the effects now, one question though.

It seems as if PrimoCache is messing with the metadata of files. I have been using Resilio Sync for a long time and never had a problem with shares getting stuck, but ever since I started using PrimoCache, I've had problems with shares being stuck on uploading files and when submitting debug logs to Resilio support I always get the same response, metadata is missing.

As quoted by support:
"Haley Webb (Sync)

Jul 4, 03:48 SST
Hello,

The problem is still in calculating metadata for the files inside the share."

[Support]

Thanks for the bug report! We're checking this problem and will follow this topic if we have some updates.