diff --git a/www/comparison.html b/www/comparison.html index 7bb3a7b60..066fb8e37 100644 --- a/www/comparison.html +++ b/www/comparison.html @@ -43,7 +43,7 @@

Features comparison between Linux SCSI targets

-

As on July 2009

+

As on August 2009

@@ -164,10 +164,14 @@ ORDERED commands + (Windows 2008 clustering / RHEL5 I/O fencing) - + @@ -417,8 +421,6 @@ reinstatement 21
+ + -,
data
Persistent Reservations
- (Windows 2008 clustering)		 - - - Under development Under development, + patch is + available for + testing - - Under development
ALUA - - - Under development ALUA - - - Under development
Failover Clustering + ? + + Safe Not s existing connections with some errors. However, your iSCSI initiator also should be able to handle the safe restart. For instance, old (pre-CentOS/RHEL 5) open-iscsi has problems in this area. But the latest versions do it pretty well.

- - diff --git a/www/comparison_history.html b/www/comparison_history.html deleted file mode 100644 index a0a0c8242..000000000 --- a/www/comparison_history.html +++ /dev/null @@ -1,126 +0,0 @@ - - - - - - - - -SCSI Targets Comparison History - - - - -
- - - - - -
- - -
- -

Some background history of the Features comparison page.

- -

LIO was the first who has had similar comparison - -. It was and still is very much -wrong about SCST. I asked that page author, Nicholas Bellinger, to -correct the wrong items (see my -e-mail), but Nicholas Bellinger refused to do it without any explanations. Moreover, he -blocked my access to the LIO mailing list, preventing me from telling that -to the interested people myself. After all our previous discussions and -with his skills and experience it's nearly impossible to believe that Nicholas Bellinger -didn't know that SCST is a lot more generic than LIO and has zero-copy -in all the same places, where LIO has. Thus, that comparison page is looking like rather a deliberate cheating -attempt. Seems SCST is so much superior over LIO, so Nicholas Bellinger gave up technical discussions and started -attacking people's perception about SCST, trying to inspire them the opposite.

- -

Update

- -

After our comparison page was published, Nicholas Bellinger corrected in his comparison page the most crying items about SCST. Namely: -he acknowledged that SCST is a generic engine as well as marked "zero-copy" item for SCST as "Needs kernel patch" and removed "User Interface" item. -But "zero-copy" mark for SCST is still wrong, because -SCST needs kernel patch to provide send zero-copy in the iSCSI target driver (and only in it) only if it is working with user space backend. LIO doesn't -support user space backend, so, if it has "X" in this line, SCST must also have unconditional "X" there.

- -

Moreover, new wrong items where introduced: - -

    - -
  • SCST was marked as "-" in "Additional Header Segment" and "Bidirectional Commands" items. - SCST and iSCSI-SCST support bidirectional commands as well as iSCSI-SCST supports other values in the - Additional Header Segment (namely, extended CDB).
    • - -
  • STGT marked as having Fibre Channel target drivers for Emulex, QLogic and LSI adapters. Where are they?
    • - -
  • STGT marked as having Fibre Channel QLogic target driver in-tree. Which tree is meant?
    • - - -
  • STGT marked as supporting SAS and parallel SCSI targets. But - it fundamentally can't do that. Those transports don't provide expecting - transfers values (size and direction), so those values must be extracted - directly from CDBs. STGT doesn't have the necessary infrastructure for that, see, for instance, - "https://lists.berlios.de/pipermail/stgt-devel/2007-March/000436.html. - Nothing has changed since time, when that e-mail was sent. - -
- -

Interesting coincidence that the above wrong STGT items were added at nearly the same time as LIO started working with STGT's in-kernel target drivers...

- -

- -
-
-
- - - - - - - - - - - -