March 2010 - WLUG - wlug.mailman3.com

Re: [Wlug] Meeting topics

by Jim Dibb

But then again, you didn't state how important your uptime of your system is. Sorry for leaving that bit out.... On Mon, Mar 1, 2010 at 4:45 PM, Jim Dibb <jimdibb(a)gmail.com> wrote: > > > On Mon, Mar 1, 2010 at 4:21 PM, Chuck Anderson <cra(a)wpi.edu> wrote: > >> >> >> I'm interested. I was considering using SSDs for my boot drive. Is >> RAID protection necessary/advisable using SSDs? >> _______________________________________________ >> Wlug mailing list >> Wlug(a)mail.wlug.org >> http://mail.wlug.org/mailman/listinfo/wlug >> > > SSDs have much higher reliability than HDD. > Mirroring is the only RAID level that does not impose a write penalty. > Mirroring SSDs is expensive > > Based on these 3 points, I'd say a backup is a much better solution than > RAID, unless you want to (R)AID-0 them into a stripe for even more insane > performance. > >

March 1, 2010

Re: [Wlug] Meeting topics

by Jim Dibb

On Mon, Mar 1, 2010 at 4:21 PM, Chuck Anderson <cra(a)wpi.edu> wrote: > > > I'm interested. I was considering using SSDs for my boot drive. Is > RAID protection necessary/advisable using SSDs? > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug > SSDs have much higher reliability than HDD. Mirroring is the only RAID level that does not impose a write penalty. Mirroring SSDs is expensive Based on these 3 points, I'd say a backup is a much better solution than RAID, unless you want to (R)AID-0 them into a stripe for even more insane performance.

March 1, 2010

Re: [Wlug] Meeting topics

by Patrick Dignan

On 03/01/2010 04:21 PM, Chuck Anderson wrote: > On Sun, Feb 28, 2010 at 06:26:31PM -0500, Jeff Moyer wrote: > >> If anyone is interested in a talk on SSDs, chime in. I can present my >> experiences with these devices. I've evaluated the performance of >> several vendors' products, ranging from the poor to the exemplary. I >> can talk about the modifications made in the block layer to accommodate >> for these devices, and the current implementation of the block discard >> support (which maps to ATA TRIM, which is important for device lognevity >> and performance) and how that is wired up to file systems such as ext4. >> > I'm interested. I was considering using SSDs for my boot drive. Is > RAID protection necessary/advisable using SSDs? > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug > I second Chuck here.

March 1, 2010

Re: [Wlug] Meeting topics

by Chuck Anderson

On Sun, Feb 28, 2010 at 06:26:31PM -0500, Jeff Moyer wrote: > If anyone is interested in a talk on SSDs, chime in. I can present my > experiences with these devices. I've evaluated the performance of > several vendors' products, ranging from the poor to the exemplary. I > can talk about the modifications made in the block layer to accommodate > for these devices, and the current implementation of the block discard > support (which maps to ATA TRIM, which is important for device lognevity > and performance) and how that is wired up to file systems such as ext4. I'm interested. I was considering using SSDs for my boot drive. Is RAID protection necessary/advisable using SSDs?

March 1, 2010

Re: [Wlug] Meeting topics

by Frank Sweetser

On 02/28/2010 06:26 PM, Jeff Moyer wrote: > If anyone is interested in a talk on SSDs, chime in. I can present my > experiences with these devices. I've evaluated the performance of > several vendors' products, ranging from the poor to the exemplary. I > can talk about the modifications made in the block layer to accommodate > for these devices, and the current implementation of the block discard > support (which maps to ATA TRIM, which is important for device lognevity > and performance) and how that is wired up to file systems such as ext4. Sounds like a good talk to me. -- Frank Sweetser fs at wpi.edu | For every problem, there is a solution that WPI Senior Network Engineer | is simple, elegant, and wrong. - HL Mencken GPG fingerprint = 6174 1257 129E 0D21 D8D4 E8A3 8E39 29E3 E2E8 8CEC

March 1, 2010