Benchmarks Of The Btrfs Space Cache Option

[madjr]

so how long you guys think will take for btrfs to become default in most distros?

and did the fedora guys (or whom was working on it) finish the grub integration for the snapshots?

[deanjo]

so how long you guys think will take for btrfs to become default in most distros?

and did the fedora guys (or whom was working on it) finish the grub integration for the snapshots?

My guess is 2 to 3 years for most of the mainstream distros. Wouldn't be surprised if *buntu jumped on it sooner though, the tend to jump on green solutions before they are ready for primetime.

[nevertell]

What about the cpu usage differences ?

[pvtcupcakes]

Have you done any long term analysis of the compress option?

I used it on both my root and home partitions for a while, at the start the speed was amazing, but slowly over the course of a few months it became slower and slower

It even started to effect the way Chrome and FireFox worked as it was taking so long to open their cache / database

Has anyone else noticed this?

I'm just worried that your praising the compress option because it's only been tested on clean installs and it's not being used for prolonged periods of time

Chrome does that for me even on my ext4 machine. I had to symlink Chrome's cache to /dev/null because it gets ridiculously huge like 1.5gb and then Chrome locks up my machine for a good 3 minutes when it starts up.

After reading this article, I should start using compress on Btrfs. Can I just enable it now, or should I reformat and start fresh. I don't know if it's good to have a mix of compressed and uncompressed.

[sabriah]

Could you compare the results with same configurations but a magnetic hard disk instead of ssd?

This may be more useful than one may think.

If one examines the "drive ready time" timings for different hard drives, this may be a major factor. Looking at http://www.google.com/search?q=hard+...dy+time%22+sec one may see that the "drive ready time" timings may vary from 5 to 25 seconds! The Western Digital Caviar Black 2TB 6GB/sec has 21 seconds, which is not good.

I can imagine that there is some relevance to how well the Space Cache Option performs at either end of that interval.

[Wyatt]

Like another person, I want CPU time numbers too. Fast I/O that takes an entire core or is pegged to my CPU speed is worthless to me (and most anyone that uses a low-power or dynamic power-state device, such as a laptop, mobile phone, set-top box, embedded device, etc.)

[madjr]

My guess is 2 to 3 years for most of the mainstream distros. Wouldn't be surprised if *buntu jumped on it sooner though, the tend to jump on green solutions before they are ready for primetime.

maybe ubuntu jumping on it could speed things up a bit?

[deanjo]

maybe ubuntu jumping on it could speed things up a bit?

With the amount of development that they do, I doubt it. To get it really rolling I imagine Fedora would have to default to it first before adoption would start picking up.

[liam]

With the amount of development that they do, I doubt it. To get it really rolling I imagine Fedora would have to default to it first before adoption would start picking up.

I assume he was referring to the bug reports that would be produced. There might be somewhat of an issue with variety of installations that the btrfs devs are receiving.

[deanjo]

I assume he was referring to the bug reports that would be produced. There might be somewhat of an issue with variety of installations that the btrfs devs are receiving.

Even then it is questionable. Without any real clear advantage over the current default there won't be much motivation to switch from a somewhat proven file system.