Quicksearch
Disclaimer
The individual owning this blog works for Oracle in Germany. The opinions expressed here are his own, are not necessarily reviewed in advance by anyone but the individual author, and neither Oracle nor any other party necessarily agrees with them.
SAN without a storage system
Friday, January 18. 2008
I had a strange idea some days ago. The new generation of our X- or T-class chassis have bays for 8 disk right now. There must be a good usage for this mass of bays. Think about it. This means you have 320 Disks in your rack. A few seconds later i remembered about a question, somebody asked at the Niagara Presentation Event some years ago in Berlin (must have been 2005): "72 GB per disk, okay ... your OS uses 4 Gig at maximum. How can i use the rest? It would be nice, if other systems can use the storage". I had no answer for it ... okay ... the standard ones ... SMB/NFS.
Well ... there is a better idea. We have an iSCSI Target in Solaris right now and COMSTAR enables Opensolaris to present FC targets. Okay, now take all this free storage, format it ZFS and create 2G sized emulated Volumes out of it and map each of it with the respective target emulation to the LAN or SAN. Whenever you need some storage, you can pick some of the several volumes you have created from all nodes with the exception of the node you work on.
By such a configuration, a rack of servers can easly be it´s own SAN without a real storage subsystem. And you can use all the unused space on all the disks, or use own disks for such a construction with 8 or more bays you have more than enough room. With 128 SPARC Threads on 2 HU or 16 x86 threads on 2 HU there should be enough spare cycles on the CPU as well.
Just an idea, but i will think about it the next days.
Well ... there is a better idea. We have an iSCSI Target in Solaris right now and COMSTAR enables Opensolaris to present FC targets. Okay, now take all this free storage, format it ZFS and create 2G sized emulated Volumes out of it and map each of it with the respective target emulation to the LAN or SAN. Whenever you need some storage, you can pick some of the several volumes you have created from all nodes with the exception of the node you work on.
By such a configuration, a rack of servers can easly be it´s own SAN without a real storage subsystem. And you can use all the unused space on all the disks, or use own disks for such a construction with 8 or more bays you have more than enough room. With 128 SPARC Threads on 2 HU or 16 x86 threads on 2 HU there should be enough spare cycles on the CPU as well.
Just an idea, but i will think about it the next days.
Comments
Display comments as
(Linear | Threaded)
Herzlichen Glückwunsch, Sie haben das Google Filesystem erfunden. Nur komplizierter.
Hmmm ... wie in der Bosch-Werbung ... 
Aber das GoogleFS ist doch shared reader/shared writer? Das kann die obige Lösung nicht.
Aber das GoogleFS ist doch shared reader/shared writer? Das kann die obige Lösung nicht.
Ob es jeder Admin toll findet, wenn "sein" Rechner als Storage mitbenutzt wird? Darueber hinaus muss man eine zusaetzliche FC-Infrastruktur aufbauen... aber iSCSI-LDOMs koennten mir gefallen...
Ja, okay ... das mit dem FC stimmt. Aber mit dem iSCSI geht schon heute. Und wenn er mosert, einfach den Admin in einem "Container stecken" ... dann sieht er nichts davon, das die Maschine noch was anderes macht ... ansonsten ist es ja eh seltener, das in einer groesseren Farm jeder rechner von einem Admin gewartet wird.
... dann sieht er nichts davon, das die Maschine noch was anderes macht ... ansonsten ist es ja eh seltener, das in einer groesseren Farm jeder rechner von einem Admin gewartet wird.
Trotzdem wiederspricht die Lösung dem Konsolidierungsgedanken. Stoarges gehören in das Zentrum einer IT-Struktur. Es ist egal wer oder was die Daten verarbeitet - wichtig ist wo sie liegen. Trotzdem erscheint mir die Lösung für bestimmte Anwendungszwecke sicherlich interessant. 
MfG
Patrick
MfG
Patrick
+1
The LKSF book
The book with the consolidated Less known Solaris Tutorials is available for download here
Web 2.0
Contact
Networking
xing.com
LinkedIn
Facebook
My photos
- mail: joerg@c0t0d0s0.org
- icq: 144000169
- Me, myself and I
- openBC
- My del.icio.us
- c0t0d0s0@twitter
- c0t0d0s0@last.fm
- wiki@c0t0d0s0
- Amazon Wishlist
Networking
xing.com
My photos
Comments
about Nanosecond
Wed, 23.05.2012 00:11
I remember this being drummed
into us during Digital Design
at Uni. It's important to cons
ider it when laying out [...]
Mon, 21.05.2012 18:04
Hello Kevin, Im not surprised
with what you are seeing or ha
ve seen when attaching a SSD t
o a USB2.0. USB3.0 helps [...]
Mon, 21.05.2012 04:44
Hi Greg,
With regards to IO
PS I have seen terrible result
s using a 60GB SATA2 SSD with
USB2.0 - USB2 really cho [...]
about ZFS Dedup Internals
Sat, 19.05.2012 09:50
There is no impact to boot/imp
ort times, as the DDT is loade
d as needed ... so the pool is
imported as fast as wit [...]
Buttons
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Germany License
Some new systems were announced a few days ago. All are Nehalem based. And all of them have some special features seperating them from the competition. But this isn´t the point of this article. I find one system the most interesting: Really long ago i
Tracked: Apr 17, 19:19