On 02/07/2013 07:58 AM, Adam Goryachev wrote: > On 07/02/13 23:53, Phil Turmel wrote: >> On 02/07/2013 07:49 AM, Adam Goryachev wrote: >>> Well, if I can re-trim daily, and have enough clean space to work for 2 >>> days, then I should never hit this problem.... Assuming it loses *that >>> much* performance.... >> >> You keep saying this, but you are only going to be trimming the >> unpartitioned space. That won't help you after the first trim. > > > Right, so if I TRIM it each night, then the next day, there is a bunch > of freshly TRIM'ed space to use up. As long as there is enough for the > day's writes, then I won't face this issue of slow writes while the SSD > is trying to do garbage collection or whatever... Trim is a per-sector property. Once trimmed, it stays trimmed until you write to *that* sector. > Or, I'm just dreaming :) You are dreaming. If you want to achieve something with trim, you would trim on your clients and hope it passes all the way down the stack through iSCSI to your LVs and then to MD and the SSDs. > Or, this isn't my problem, so I should just ignore it.... You just have nothing useful on the server side for trim to do. (Although you should manually trim the unpartitioned space. You only need to do so once.) Phil -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
