NFS / NAS/SAN /CIFS

BTW: with "SAN is a network device that consists of storage that is directly attached to the server", I mean: directly attached or attached via it's own private network ( = storage area network ).

Nicely explained joachim.claeys@teleatlas.com but please also provide links to text used from other sites to protect EE from copyright infringements:
https://www.experts-exchange.com/help.jsp#hi125
https://www.experts-exchange.com/help.jsp#hi60

http://searchwinit.techtarget.com/sDefinition/0,,sid1_gci214121,00.html
http://searchstorage.techtarget.com/sDefinition/0,,sid5_gci214410,00.html
etc.
Thanks.

At the bottom of the "theory" text, I mention "whatis.com". Maybe you had overlooked it.

Very sorry joachim.claeys@teleatlas.com, yes, I did miss that. As mentioned, nicely explained, and it is common and quite acceptable to pull info from other sites, but though I should point out "the rules" where you are relatively new to the forum. Obviously you are aware. Guess I didn't get to the very bottom. The links I saw were from all different sites, so I assumed they were missing, they must be all linked from whatis.com
Again I apologize, and I appreciate your tactful " Maybe you had overlooked it." <G>.
Cheers!
--Rob

I don't understand when they say with SAN and/or NAS  clients can see the drives as they were local disks.

I guess any mapped drive will look as a local drive to the user.
Unless if the meaning here is that user can format/partion(if the user has rights) the same way he can do for the local drives.

Regarding NAS and SAN , the only difference I can see is the speed. is there anything else?

Generally, with a SAN device, you will need the required hardware to be able to access it ( a hostbus adapter or HBA, ATM card, ... ) .  On a SAN, the storage will be seen as "local" storage to your OS, because it is physically attached.
Maximum throughput depends on the technology used to connect, but is generally faster than a NAS

With a NAS device, you would typically only need a generic network interface to be able to access your files. Typically your server will look at the storage as being "network" storage ( CIFS / NFS ) e.g. A network mapped drive in the case of Windows or an NFS share in the case of UNIX, Linux based systems.
Depending on your network speed, it's bound to be slower than a SAN.

Regards,
J.

<<<<On a SAN, the storage will be seen as "local" storage to your OS, because it is physically attached>>>>>
there should be a wire (FC) running from the server to disk storage, and it should be the same for NAS that has a wire to run from the server to disk storage.
Can you clear it up?

thanks

the difference is the way you use the disk:

for SAN, using iSCSI, you can actually use standard GE NIC's, no HBA or so required. When "mounting" a disk in a SAN environment, it appears as locally attached, meaning as if it was hooked up to the SCSI controller of your motherboard. There's no permissions at this level, it is raw disk space that you can format as whatever you want and even partition. This is mainly used for business applications, ERP's, CRM's etc. because of reliability, multipathing, redundancy etc.

NAS is what you want to use in order to manage shares and their data from a sinlge point. The filesystem, formatting etc. is up to the NAS file server in this case, so it is quite different indeed.

hope this helps,

LePiaf