Measurable improvements with 1TB of Cache

I added some cache drives just to see what it would do, and the numbers show a very positive improvements

Not sure what the daily shark fins are, but the node is definitely a lot more active now.

Graphs shown are in order of week, day, month

Is this data from the nodes, or from the cache? Also how has it affected your success rate? Did you implement it on the 11th?

The view is from the Synology storage array and I don’t know how to check the success rate if it’s not in the dashboard.

Success rate script - Now updated for new delete terminology since v1.29.3 its a command line script. It shows how many “races” you won. Which I believe, unless you are losing upload(to customer) races, a read cache wouldnt help much. But at this point you would have to turn the cache off, let it run for a while to get a baseline rate, then turn it back on to see the cached rate. What HDDs do you have and how much data?

1 Like

Sorry for the delay, and thanks for the script :slight_smile:
It takes forever to run with logfiles that have more than 1861586 lines

Anyways, I’m using WD Reds (4x6TB) connected via iSCSI

Here’s the stats from a few days of Storj + Cache

611 failed (0.14%)
23309 Cancelled (5.3%)
415773 Successful (94.56%)

230 failed (0.08%)
1742 Cancelled (0.62%)
280847 Successful (99.3%)