speed

speed

[folkert]

Hi,

I've got a couple of questions regarding bcache:

- will it be faster to implement bcache instead of letting the
  write-intent-map of a raid1 volume be written on an ssd together with
  the journals of the filesystems on that raid1 volume?

- if i have multiple logical volumes in a volume group, do I need to
  explicitly configure bcache for each of them? or can I "connect" bcache
  to the whole volume group?

- does bcache do trim? and/or pass it through?

- maybe bcache can check if a written block is all 0x00 and then do a
  trim instead? that is not only usefull for SSDs but can help
  data-deduplication software/appliances as well

- what will happen if an SSD suddenly stops working in the sense that it
  won't allow anymore writes? will bcache then still commit all the
  buffered data which was already written to the SSD but not to the HDDs?

- any suggestions what SSD to buy? e.g. an OCZ Vertex 4?

- are there any suggestions on what size SSD to use for a certain size
  HDD? or does it depend on the amount of data written per second? or
  the number of iops?

- is it possible to do an emergency flush where everything in the cache
  is written to disk, *as much as possible*? e.g. if a read fails,
  continue with other blocks?
  I'm asking this as I'm not entirely convinced of the quality of SSDs

- is it possible to use more than 1 SSD as a bcache for the same
  storage? not as raid0 as you then can't remove one from the system
  without affecting the other

- does bcache work with e.g. iscsi/nbd remote storage?
  - then maybe it would be nice if it is possible to temporarily bring
    down the remote-storage + bcache combination when the network
    connection goes down

- is it possible to swap via bcache?

- maybe zcache can be combined so that you can work with more data with
  a smaller SSD?

- I think I read on the website that blocks are stored in sorted order
  on disk. that's probably on the HDD below it and not on the SSD,
  right? (because afaik an ssd has always the same access latency,
  independent of write order. altough I can image if that is not true)

- does it keep track of checksums/crcs of the data it writes to the SSD?
  e.g. it writes a block to the SSD and then later on it reads that
  block back from the SSD (to write to e.g. HDD) and verifies that what
  it got back is what it expected


Folkert van Heusden

-- 
Always wondered what the latency of your webserver is? Or how much more
latency you get when you go through a proxy server/tor? The numbers
tell the tale and with HTTPing you know them!
                                     http://www.vanheusden.com/httping/
-----------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com

Re: speed

[matthew patton]

> - any suggestions what SSD to buy? e.g. an OCZ Vertex 4?

OCZ is IMO a troubled company and their products are seriously hit/miss. I would go with Intel, samsung, micron/corsair in that order.

> - are there any suggestions on what size SSD to use for a
> certain size

It all depends on your workload. The "big boys" generally put at least 5% of backend capacity in SSD. For normal singleton server/desktops you'd be hard-pressed to see any value. If you're running virtualized servers, the combined I/O patterns seen at the central array degenerate to random I/O and that is where tiering or Bcache can be useful. The individual server running it's virtualized workload might see some benefit too but probably not to near the same extent. If you're using NFS datastores running FS-cache on the virtual host might be a very useful thing but only if the local SSD/fast HD is big enough to grab the entire file since virtual disk files tend to be sizable.

> - is it possible to use more than 1 SSD as a bcache for the
> same   storage?

use LVM segments. ie. put <N> SSDs in a VGpool and then when allocating the LV, specify extent ranges out of some/all of them. When you need to remove the SSD, use pvchange to have LVM vacate the extents on that drive. Then you can remove at will.

> - does bcache work with e.g. iscsi/nbd remote storage?
>   - then maybe it would be nice if it is possible to
> temporarily bring
>     down the remote-storage + bcache combination
> when the network
>     connection goes down

If your iSCSI is unreachable your I/O will stall and timeout regardless of any local Bcache. Oh you might get lucky for a couple seconds but after that, boom. Engineer your network-attached storage correctly.

> - is it possible to swap via bcache?

swap is generally a continuous write, so no. If you're swapping, you need to fix the underlying problem, not play games with trying to make swap "faster". You can ofcourse mount your swap device from something that is locally attached and cut out the bandwidth and latency of an over-the-wire target.

> - does it keep track of checksums/crcs of the data it writes
> to the SSD?
>   e.g. it writes a block to the SSD and then later on
> it reads that
>   block back from the SSD (to write to e.g. HDD) and
> verifies that what
>   it got back is what it expected

No. If you want ZFS' features use that. This whole silent block corruption thing is a marketing gimmick by Sun because at some point 10+ years ago they used a family of REALLY buggy controllers. Oh sure, if you've got petabytes on your petabytes you'll run into it eventually. If you use crap hardware with half-baked drivers or shitty consumer hardware you may run into it too. But so you know, the PCIe busses use checksums and erasure-coding. As does the SATA/SAS chip when sending it down the wire to the drive interface. And the drive itself does checksums and in fact writes the data using erasure coding too. So about the only time that 512bytes is just 512bytes of "fragile", unprotected data is when it's in RAM be it system, controller, or disk. Not to say that bit-flips don't happen where the drive miss-writes or miss-reads but the occurance is vanishingly rare and the drive already has mechanisms to deal with it.