I'm a member of
Quicksearch
Codenews Search
Disclaimer
The individual owning this blog works at Sun Microsystems GmbH in Germany, a subsidiary of Oracle. 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.
Navigation
Categories
About this blog- Apple
- iPhone
- Aviation
- Bahn
- Begriffe des Lebens
- Blogosphere
- Braindump
- Business Travel
- CANBOX
- CeBIT
- CEC
- Computing
- del.icio.us
- Fundsache
- General
- HHOSUG
- i hate sundays
- Kochen
- Mindfuck
- Movies
- Music
- On Benchmarketing
- Persoenliches Pech
- Photographie
- policy of ...
- Privacy
- Recommended Blogs
- Security
- Solaris
- Strange Days
- Sun/Oracle
- suncec2007
- suncec2008
- Technology
- The IT Business
- Toolbox
- Travel
- Work
Sun Storage F5100 officially announced
Monday, October 12. 2009
Sun has now announced the F5100 officially. The specs are really impressive: The largest version with 80 FMods yields 1.6 Million read IOPS and 1.2 million write IOPS with 4k blocks. 12.8 GB/sec read and 9.7 GB/sec write. 1,920 GB capacity. It uses 386 watts while active. More specs are available at the Sun specs page for the Sun Storage F5100 Flash Array. Furthermore the documentation of this devices should reappear soon at docs.sun.com. Until that moment you will find some additional information about the device at Sun's website. This device isn't that expensive as you may think. The device starts at $45,995 (with 20 Flash Modules giving you 480 GB, 397 K IOPS read and 304 K IOPS write).
When you look in such a device you will see 80 densly packed FMods.
In the back of the device you will find the SAS connectors for the device. There are four domains with 4 SAS connectors each. For a maximum performance configuration you configure 5 FMods and a SAS connector into a zone.
Most important: The FMods doesn't have a battery on each module to protect the caches. This is done centrally by super capacitors in so called Energy Storage Modules. You plug them into the front of these devices.
Just a warning: Those ESM are quite potent super caps. Read the manual with care before handling those modules. You need a lot of power to protect all the caches and those ESM are capable to deliver this power.
When you look in such a device you will see 80 densly packed FMods.
In the back of the device you will find the SAS connectors for the device. There are four domains with 4 SAS connectors each. For a maximum performance configuration you configure 5 FMods and a SAS connector into a zone.
Most important: The FMods doesn't have a battery on each module to protect the caches. This is done centrally by super capacitors in so called Energy Storage Modules. You plug them into the front of these devices.
Just a warning: Those ESM are quite potent super caps. Read the manual with care before handling those modules. You need a lot of power to protect all the caches and those ESM are capable to deliver this power.
Comments
Display comments as
(Linear | Threaded)
9.7 GB/sec IOPS => 9.7 GB/sec write?
Looks pretty, but the wiring limitations in the manuals was a bit perplexing; couldn't connect a domain to more than one host at a time. so why put 4 ports per domain?
Looks pretty, but the wiring limitations in the manuals was a bit perplexing; couldn't connect a domain to more than one host at a time. so why put 4 ports per domain?
You can zone a domain into four groups and put a SAS port in each of this zones. You need those ports to get the most out of a domain because a single SAS 4x isn't capable to drive it.
Thanks for the hint. I've corrected the typo with the GBps IOPS ...
There are a lot of single point of failure. For FMod replacement you have to stop storage.
Also you can`t monitor FMods errors.
It will be a good device for HPC if it will have Infiniband ports, but for database storage its very poor architecture.
And I also think, that throughput of servers HBAs is smaller than this enclosure.
May be Alpha server can provide this throughput.
Also you can`t monitor FMods errors.
It will be a good device for HPC if it will have Infiniband ports, but for database storage its very poor architecture.
And I also think, that throughput of servers HBAs is smaller than this enclosure.
May be Alpha server can provide this throughput.
1. You can connect up to 16 HBA to this device.
2. For databases you would use 2 of this devices in a mirrored configuration. It's not meant as a single storage device.
3. When used as L2ARC or SZIL you would have multiple points of getting slower, not of failure (of course sometimes it's the same), as ZFS would switch over to using normal ZIL or don't use L2ARC.
4. Health-checking FMods and ESM is done via the Sun Common Array Manager
2. For databases you would use 2 of this devices in a mirrored configuration. It's not meant as a single storage device.
3. When used as L2ARC or SZIL you would have multiple points of getting slower, not of failure (of course sometimes it's the same), as ZFS would switch over to using normal ZIL or don't use L2ARC.
4. Health-checking FMods and ESM is done via the Sun Common Array Manager
1. You can connect up to 16 HBA to this device.
You have limited number of IO slots per server, and you also need connect minimum 4 ports if you use full loaded F5100. If you use dual path you need 8 connections to DB server.
4. Health-checking FMods and ESM is done via the Sun Common Array Manager
I`m pretty shore, that this can`t monitor life time of FMod, cause FMod will have only performance decrement.
You have limited number of IO slots per server, and you also need connect minimum 4 ports if you use full loaded F5100. If you use dual path you need 8 connections to DB server.
4. Health-checking FMods and ESM is done via the Sun Common Array Manager
I`m pretty shore, that this can`t monitor life time of FMod, cause FMod will have only performance decrement.
In max performance configuration, this system isn't meant to be used by a single server or when used by a single server to be used by a large one that has the capability to drive 1.6M IOPS like the M9000 class.
For configurations where you need more than the available slots, you could use the External I/O Unit (http://www.sun.com/servers/midrange/external_io/features.xml) which is supported by all T5xxx and M4000 to M9000.
As described by the manual multipathing is not supported (http://docs.sun.com/source/835-0772-01/cggfgcfb.html). For data availability with a single box you would mirror domains.
For configurations where you need more than the available slots, you could use the External I/O Unit (http://www.sun.com/servers/midrange/external_io/features.xml) which is supported by all T5xxx and M4000 to M9000.
As described by the manual multipathing is not supported (http://docs.sun.com/source/835-0772-01/cggfgcfb.html). For data availability with a single box you would mirror domains.
Links in this article
- Sun Storage F5100 Flash Array
- docs.sun.com
- https://www.c0t0d0s0.org/uploa
ds/k3_flash-f5100_6.jpg - https://www.c0t0d0s0.org/uploa
ds/k3_flash-f5100_4.jpg - https://www.c0t0d0s0.org/uploa
ds/k3_flash-f5100_2.jpg - Sun's website
- http://www.c0t0d0s0.org/upload
s/k3_flash-f5100_6.jpg - http://www.c0t0d0s0.org/upload
s/k3_flash-f5100_4.jpg - http://www.c0t0d0s0.org/upload
s/k3_flash-f5100_2.jpg
The LKSF book
The book with the consolidated Less known Solaris Tutorials is available for download here
Twitterfeeds
twitter.com/c0t0d0s0
@mperedim no ... research for a new blog article ;)
twitter.com/codenews
6914386 X freeze (and reboot) a build 130 system http://bit.ly/abvIH5
twitter.com/SunPatches
Security patch: 113723-21 - SE3510 423A: StorEdge 3510 array controller firmware upgrade. Available since Feb/08/10. http://bit.ly/btnK9U
twitter.com/SolPatchesX86
109810-12 - SunOS 5.8_x86: timezone data patch. Available since Feb/08/10. http://bit.ly/bW5k68
twitter.com/SolPatchesSPARC
109809-12 - SunOS 5.8: timezone data patch. Available since Feb/08/10. http://bit.ly/cNUNg8
@mperedim no ... research for a new blog article ;)
twitter.com/codenews
6914386 X freeze (and reboot) a build 130 system http://bit.ly/abvIH5
twitter.com/SunPatches
Security patch: 113723-21 - SE3510 423A: StorEdge 3510 array controller firmware upgrade. Available since Feb/08/10. http://bit.ly/btnK9U
twitter.com/SolPatchesX86
109810-12 - SunOS 5.8_x86: timezone data patch. Available since Feb/08/10. http://bit.ly/bW5k68
twitter.com/SolPatchesSPARC
109809-12 - SunOS 5.8: timezone data patch. Available since Feb/08/10. http://bit.ly/cNUNg8
Web 2.0
Contact
Networking
open.bc
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
open.bc
My photos
Syndication
Tagged articles
AMD Apple avs Bahn Blogging Blogosphere braindump Business Travel CeBIT cec cec2006 CMT del.icio.us deutsch dtrace fliegen Fundsache General Hamburg IBM i hate sundays Intel iscsi jumpstart Links Linux lksf Mindfuck Movies Music Musik Niagara Opensolaris Opteron Photographie policy of ... Politik Security Solaris storage Sun suncec2007 sunw t1 The IT Business Ultrasparc ultrasparc t1 Wirtschaft Work ZFS
Comments
Tue, 09.02.2010 11:44
Is there anything this comment
should tell me ?
Tue, 09.02.2010 11:40
Dedup your brain!
Tue, 09.02.2010 11:25
Interesting read and it inspir
ed me to check upon the dedupe
features in TSM6.1. Seems tha
t it uses SHA-1, non-com [...]
Tue, 09.02.2010 09:32
ZFS computes the checksums any
way. The difference with hash-
only dedup is just the lookup
to a table, with hash-an [...]
Tue, 09.02.2010 09:04
Interesting Document, especial
ly Page three and four.
I c
an't agree that Support over I
nternet Portalsl/Metalin [...]
Buttons
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Germany License
Tracked: Oct 12, 15:44
Tracked: Oct 14, 05:09