All of lore.kernel.org
 help / color / mirror / Atom feed
* New to working with stgt
@ 2014-01-23  0:52 Steven Graf
  2014-01-24  5:19 ` FUJITA Tomonori
  0 siblings, 1 reply; 2+ messages in thread
From: Steven Graf @ 2014-01-23  0:52 UTC (permalink / raw)
  To: stgt

[-- Attachment #1: Type: text/plain, Size: 711 bytes --]

Hello Everyone,

I am new to the list and I have recently just started working with the
stgt framework.  I've been grokking the code and the documentation, but
I'd appreciate any insight or advice you may be able to lend.  My
objective is to use the rbd backend, but have a user-space driver that
speaks native SCSI (block device).  I am looking at the existing
implementation that uses iSCSI, and I have that all working with my Ceph
back-end, but for performance reasons I'd like to eliminate the overhead
of the network (currently I have everything going over localhost, which
is good for a PoC, but insufficient for our big data tasks).

Thanks!
-- 
Steven Graf
President
Resolutech, LLC


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 3746 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: New to working with stgt
  2014-01-23  0:52 New to working with stgt Steven Graf
@ 2014-01-24  5:19 ` FUJITA Tomonori
  0 siblings, 0 replies; 2+ messages in thread
From: FUJITA Tomonori @ 2014-01-24  5:19 UTC (permalink / raw)
  To: sgraf; +Cc: stgt

Hi,

On Wed, 22 Jan 2014 19:52:23 -0500
Steven Graf <sgraf@resolutech.net> wrote:

> I am new to the list and I have recently just started working with the
> stgt framework.  I've been grokking the code and the documentation, but
> I'd appreciate any insight or advice you may be able to lend.  My
> objective is to use the rbd backend, but have a user-space driver that
> speaks native SCSI (block device).  I am looking at the existing

I'm lost here, user-space driver speaking native SCSI? What SCSI
transport protocol you use?

> implementation that uses iSCSI, and I have that all working with my Ceph
> back-end, but for performance reasons I'd like to eliminate the overhead
> of the network (currently I have everything going over localhost, which
> is good for a PoC, but insufficient for our big data tasks).

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-01-24  5:19 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-01-23  0:52 New to working with stgt Steven Graf
2014-01-24  5:19 ` FUJITA Tomonori

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.