Jump to content
RESET Forums (homeservershow.com)

SSD Setup Practices by Forum Users


ImTheTypeOfGuy
 Share

Recommended Posts

  • Replies 64
  • Created
  • Last Reply

Top Posters In This Topic

  • ImTheTypeOfGuy

    7

  • pcdoc

    9

  • ikon

    24

  • c0op3r

    8

Go ghetto like I did add Intel Quad Nics and Use teaming

 

But you have to have compatible switch(es). So much Ethernet link bonding is proprietary ATM.

Link to comment
Share on other sites

  • 2 months later...

I recently installed a 240 GB OCZ SSD and changed my strategy for SSD changes a bit. See the first post for the changes.

Link to comment
Share on other sites

  • 2 months later...

But you have to have compatible switch(es). So much Ethernet link bonding is proprietary ATM.

Linux has some round-robin like implementations of port bonding that work well on any switch.

 

Good luck running a bridge on it though.

Link to comment
Share on other sites

Also, since the information is hard to find all in one place, here are some good practices for using SSDs and linux.

  • Format all filesystems ext4 as it supports discards (TRIM)
  • Edit /etc/fstab and replace "default" with discard,noatime. Only add the 'discard' option for SSD mount points.
  • If you have a SSD with a slow write speed, or want faster responsiveness, add "vm.swappiness = 20" to your /etc/sysctl.conf file. This will cause your system to use more memory, as it wont swap from memory to swapspace as much. (Note the default "swappiness value is 60)
  • If you use LVM (as you should) make sure you add the discard option in /etc/lvm/lvm.conf
  • If this is for a laptop, or battery-backed server (or you are working with non critical data), play around with the 'commit' mount option. This will change how often the kernel flushes writes from the write-back cache to your disk. The default is 5 seconds. Most people use (depending on how confident they are with their system) between 5 and 600 seconds (10 minutes). I prefer to use commit=300 (5 minutes) as I don't really have a write overhead on my system (all virtual machines have noatime set, and only have 1 windows VM). Note that when you perform certain operations or shutdown/restart your system the writeback cache will flush to disk. You can also type 'flush' into your shell and manually start a cache flush that way. However, if you set commit=600 as an option in /etc/fstab and your system crashes 7 minutes after your last flush, you will loose 7 minutes worth of data. Also, this too will cause higher memory usage under normal circumstances.

 

I hope that wasn't to long winded, in either case, happy computing.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share


×
×
  • Create New...