You are not logged in.
The Arch wiki mentions that when partitioning an SSD you need to align the partitions to the erase block size (see https://wiki.archlinux.org/index.php/Pa … ate_drives ). There are a lot of how-to-partition-an-ssd-under-linux which also mention this value, even an online calculater.
Those value are known for a lot of the serious models, among them also the Samsung 840 Pro and EVO. As I already asked in an older thread, I would like to know the value for the current model Samsung 850 Pro. Unfortunately Samsung Germany's support refuses to provide me with those values.
Therefore I ask myself, is it still necessary to align partititions to a multiple of the EBS, or is this unnecessary if the device supports TRIM?
Thanks!
Offline
Did you read the section immediately after the one you linked? https://wiki.archlinux.org/index.php/Pa … ng_tools_2
In past, proper alignment required manual calculation and intervention when partitioning. Many of the common partition tools now handle partition alignment automatically:
...
To verify a partition is aligned, query it using /usr/bin/blockdev as shown below, if a '0' is returned, the partition is aligned
Is that not sufficient for your purposes?
Last edited by 2ManyDogs (2014-11-04 18:55:37)
Offline
I'm not sure about the following paragraph in the wiki. Obviously different SSD models have very different EBS value (840 Evo vs. 840 Pro = 1024 KiB vs. 1536 KiB) those tools would need to know the value for a specific model, too. I doubt that the values for the 850 Pro are compiled in one of the partition tools. Gdisk defaults to
> On new disks, GPT fdisk attempts to align partitions on 2048-sector (1MiB)
> boundaries by default, which optimizes performance for all of these disk types.
The 840 EVO e.g. has an EBS of 1536 KiB, the 840 Pro on of 1024 KiB.
Offline
Reads/writes are aligned on page size, and erases happen in the background when the SSD is idle. As such, erase block size does not matter. Page size matters, but it's always small so just do MiB-alignment. The filesystem will break it down to something around 4k anyway. It's not possible to align to 1536KiB, not with hundreds of small files in a filesystem. Even if you were to put your partitions on 1536KiB boundaries and set your filesystem up with a 1536KiB raid stride, you would not notice any difference. Reads/writes are done on page level, and redistributed anyway by a flash translation layer. So no matter how much you try to align from the outside, the SSD will mangle it to its own purposes anyway, so what's the point?
Offline
Frostschutz, thanks for the clarification, this make things a lot easier of course. I'll update the wiki page accordingly.
Offline