34461055-Storage-Basics

8/7/2019 34461055-Storage-Basics

1/8

DAS-SAN-NAS

Direct-attached storage (DAS) refers to a digital storage system directly attachedto a server or workstation, without a storage network in between.

Network-attached storage (NAS) is file-level computer data storage connected toa computer network providing data access to heterogeneous network clients.

A NAS unit is essentially a self-contained computer connected to a network, with thesole purpose of supplying file-based data storage services to other devices on thenetwork.

NAS uses file-based protocols such as NFS (popular on UNIX systems), SMB/CIFS(Server Message Block/Common Internet File System) (used with MS Windowssystems), or AFP (used with Apple Macintosh Computers. NAS units rarely limitclients to a single protocol.

NAS provides both storage and filesystem. This is often contrasted with SAN(Storage Area Network), which provides only block-based storage and leavesfilesystem concerns on the "client" side. SAN protocols are SCSI, Fibre Channel,iSCSI, ATA over Ethernet (AoE), or HyperSCSI.

A storage area network (SAN ) is an architecture to attach remote computerstorage devices (such as disk arrays, tape libraries, and optical jukeboxes) to serversin such a way that the devices appear as locally attached to the operating system.Although the cost and complexity of SANs are dropping, they are still uncommonoutside larger enterprises.

Network attached storage (NAS), in contrast to SAN, uses file-based protocols suchas NFS or SMB/CIFS where it is clear that the storage is remote, and computersrequest a portion of an abstract file rather than a disk block.

Figure1 for reference:

Figure2 for reference:


2/8


3/8

to each port. Major HBA manufacturers are Brocade Communications Systems,Emulex, QLogic, LSI, Sun StorageTek, Agilent Technologies, and ATTO Technology.There are HBA models of different speeds: 1Gbit/s, 2Gbit/s, 4Gbit/s, 8Gbit/s,10GBit/s and 20GBit/s.

Fiber Channel Ports

Fiber Channel uses a shorthand terminology to describe different types ofconnections to the fiber channel network.

Fiber Channel uses the term "ports" and defines seven different types of ports:

ShortName

DescriptiveName

DeviceType

Port Function

N-port Network Port NodesNode port used to connect a node to a FiberChannel switch

F-port Fabric Port Switches Switch port used to connect the FiberChannel fabric to a node

L-port Loop Port NodesNode port used to connect a node to a FiberChannel loop

NL-portNetwork + LoopPort

NodesNode port which connects to both loops andswitches

FL-portFabric + LoopPort

SwitchesSwitch port which connects to both loopsand switches

E-port Extender Port SwitchesUsed to cascade fiber channel switchestogether

G-port General Port SwitchesGeneral purpose port which can beconfigured to emulate other port types

What is a World Wide Name (WWN)?

A World Wide Name, or WWN, is a 64-bit address used in fibre channel networks touniquely identify each element in a Fibre Channel network.

The use of World Wide Names for security purposes is inherently insecure, becausethe World Wide Name of a device is a user-configurable parameter.

For example, to change the World Wide Name (WWN) of an Emulex HBA, the userssimply needs to run the `elxcfg` command.

What is SAN zoning?

SAN zoning is a method of arranging Fibre Channel devices into logical groups overthe physical configuration of the fabric.

Each device in a SAN may be placed into multiple zones.


4/8

What are hard and soft zoning?

Hard zoning is zoning which is implemented in hardware. Soft zoning is zoning whichis implemented in software.

Hard zoning physically blocks access to a zone from any device outside of the zone.

Soft zoning uses filtering implemented in fibre channel switches to prevent portsfrom being seen from outside of their assigned zones. The security vulnerability insoft zoning is that the ports are still accessible if the user in another zone correctlyguesses the fibre channel address.

What is port zoning?

Port zoning utilizes physical ports to define security zones. A users access to data isdetermined by what physical port he or she is connected to.

With port zoning, zone information must be updated every time a user changesswitch ports. In addition, port zoning does not allow zones to overlap.

What is WWN zoning?

WWN zoning uses name servers in the switches to either allow or block access toparticular World Wide Names (WWNs) in the fabric.

A major advantage of WWN zoning is the ability to recable the fabric without havingto redo the zone information.

WWN zoning is susceptible to unauthorized access, as the zone can be bypassed if anattacker is able to spoof the World Wide Name of an authorized HBA.

Emulex Software products

? HBAnyware, centralized HBA management utility supporting Windows, Linux,Solaris and VMware ESX.

? AutoPilot Manager, HBA management utility for SMBs? AutoPilot Installer? VMPilot, Virtual machine and NPIV management utility for use with Microsoft

Virtual Server? lpfc Emulex Linux Drivers for HBAs

qlogic Software products

? SANsurfer Management Suite:o SANsurfer FC HBA Managero SANsurfer iSCSI HBA Managero SANsurfer Switch Managero SANsurfer Command Line Interface (CLI), for Windows, Linux and

Solaris? PathScale Compiler Suite (C, C++, and Fortran95)? InfiniPath MPI

All devices on a storage area network need to be uniquely identifiable.


5/8

The SAN equivalent of a MAC address is the worldwide name (WWN). You will alsosee references to WWPN (Worldwide Port Name) and WWNN (Worldwide NodeName). Usually, the WWPN is the name you will more likely use as it is the name anHBA will present when logging into a SAN fabric. Worldwide Names consist of 16hexadecimal digits grouped as 8 pairs. These are written with colon charactersseparating each pair. Some WWN examples are shown below:

50:06:04:81:D6:F3:45:42 (EMC disk subsystem)10:00:00:00:c9:22:fc:01 (Emulex HBA card)21:00:00:e0:8b:05:05:04 (Qlogic HBA card)

The format of the WWN is defined by the IEEE, The Institute of Electrical andElectronics Engineers, a global (non-profit) organization responsible for setting andmaintaining a large number of IT standards. You can find full details of the WWNstandard from the IEEE but heres a quick breakdown.

The specific format of a WWN is defined by the Name Address Authority field value,which is the first digit of the WWN. The following values are used:

1 IEEE 803.2 standard 48 bit ID2 IEEE 803.2 extended 48-bit ID5 IEEE Registered Name6 IEEE Extended Registered Name

Most commonly seen are options 1, 2 and 5. We will not discuss format 6.

IEEE Standard (NAA=1)

This format is divided into 3 sections. We will use the previous Emulex exampleabove to illustrate this format.

Section 1identifiesthe WWNas a

standard format WWN. Only one of the 4 digits is used, the other three must be zerofilled. Section 2 is called the OUI or company_id and identifies the vendor (more onthis later). Part 3 is a unique identifier created by the vendor.

IEEE Extended (NAA=2)

This format is divided into 4 sections. Using the Qlogic example, we can illustrate

this:

Section 1identifiesthe WWNas an

extended format WWN. Section 2 is a vendor specific code and can be used toidentify specific ports on a node or to extend the serial number (section 4) of theWWN. Section 3 identifies the vendor. Section 4 is the unique vendor-supplied serialnumber for the device.

IEEE Registered Name (NAA=5)

10:00 00:00:c9 22:fc:01

Section 1 Section 2 Section 3

2 0:00 00:e0:8b 05:05:04

Section 1 Section 2 Section 3 Section 4


6/8

Format 5 enables vendors to create unique identifiers without having to maintain adatabase of serial number codes. This format tends to be used by hardware vendorsto create unique WWN values based on the serial number and physical port numberof hardware devices such as disk arrays. The format has 3 sections, using the EMCexample we can illustrate this.

Section 1identifiesthis as aregistered

name WWN. Section 2 identifies the vendor. Section 3 is a vendor-specific generatedcode, usually based on the serial number of the device, such as a disk subsystem.The hardware vendor will use an algorithm to generate the WWN. The EMC algorithmis quite complex and for disk subsystems is based on the array serial number andthe port number. The code demonstrated above represents a WWN for port 3aa on adisk frame with serial number 123456789.

Company Ids

All vendors wishing to create WWNs must register for a company ID or OUI(Organizationally Unique Identifier). These are maintained and published by IEEE,the current list can be found here.

All company identifiers are six digits in length and used within the WWN. In ourexamples, 00-60-48 is the code for EMC, 00-00-C9 is the code for Emulex and 00-E0-8B is the code for Qlogic.

What exactly is i-SCSI?

In computing, iSCSI (pronounced or eye-scuzzy), is an abbreviation ofInternet

Small Computer System Interface, an Internet Protocol (IP)-based storagenetworking standard for linking data storage facilities. By carrying SCSI commandsover IP networks, iSCSI is used to facilitate data transfers over intranets and tomanage storage over long distances. iSCSI can be used to transmit data over localarea networks (LANs), wide area networks (WANs), or the Internet and can enablelocation-independent data storage and retrieval. The protocol allows clients (calledinitiators) to send SCSI commands (CDBs) to SCSI storage devices (targets) onremote servers. It is a popular storage area network (SAN) protocol, allowingorganizations to consolidate storage into data center storage arrays while providinghosts (such as database and web servers) with the illusion of locally-attached disks.Unlike traditional Fibre Channel, which requires special-purpose cabling, iSCSI can berun over long distances using existing network infrastructure.

Storage array

In a data center or enterprise environment, an iSCSI targetoften resides in a large storage array, such as a NetApp filer oran EMC Corporation NS-series computer appliance. A storagearray usually provides distinct iSCSI targets for numerousclients.

5 0:06:04:8 1:D6:F3:45:42

Section 1 Section 2 Section 3


7/8

Must Read:

Notable families of disk interfaces include:

? Historical bit serial interfaces connect a hard disk drive (HDD) to a harddisk controller (HDC) with two cables, one for control and one for data. (Eachdrive also has an additional cable for power, usually connecting it directly tothe power supply unit). The HDC provided significant functions such asserial/parallel conversion, data separation, and track formatting, and requiredmatching to the drive (after formatting) in order to assure reliability. Eachcontrol cable could serve two or more drives, while a dedicated (and smaller)data cable served each drive.

o ST506 used MFM (Modified Frequency Modulation) for the dataencoding method.

o ST412 was available in either MFM or RLL (Run Length Limited)encoding variants.

o Enhanced Small Disk Interface (ESDI) was an interface developed byMaxtor to allow faster communication between the processor and the

disk than MFM or RLL.

? Modern bit serial interfaces connect a hard disk drive to a host businterface adapter (today typically integrated into the " south bridge") with onedata/control cable. (As for historical bit serial interfaces above, each drivealso has an additional power cable, usually direct to the power supply unit.)

o Fibre Channel (FC), is a successor to parallel SCSI interface onenterprise market. It is a serial protocol. In disk drives usually theFibre Channel Arbitrated Loop (FC-AL) connection topology is used. FChas much broader usage than mere disk interfaces, it is thecornerstone ofstorage area networks (SANs). Recently other protocolsfor this field, like iSCSI and ATA over Ethernet have been developed as

well. Confusingly, drives usually use copper twisted-pair cables forFibre Channel, not fibre optics. The latter are traditionally reserved forlarger devices, such as servers or disk array controllers.

o Serial ATA (SATA). The SATA data cable has one data pair fordifferential transmission of data to the device, and one pair fordifferential receiving from the device, just like EIA-422. That requiresthat data be transmitted serially. Similar differential signaling systemis used in RS485, LocalTalk, USB, Firewire, and differential SCSI.

o Serial Attached SCSI (SAS). The SAS is a new generation serialcommunication protocol for devices designed to allow for much higherspeed data transfers and is compatible with SATA. SAS uses amechanically identical data and power connector to standard 3.5"SATA1/SATA2 HDDs, and many server-oriented SAS RAID controllers

are also capable of addressing SATA hard drives. SAS uses serialcommunication instead of the parallel method found in traditional SCSIdevices but still uses SCSI commands.


8/8

List of Fibre Channel sw itches

Major manufacturers ofFibre Channel switches are: Brocade, Cisco, McData andQLogic.

? Brocade:o Switches: 5300, 5100, 5000, 4900, 2400, 2800, 3800, 3900, 4100,

300, 200Eo Directors: 12000, 24000, 48000 and DCX Backboneo More complete list in Brocade Communications Systems article.

? Cisco:o Switches: Cisco MDS 9016, 9020, 9032, 9112, 9120, 9124, 9124e,

9134, 9140, 9148, 9216, 9216i, 9222i, 9302, 9304, 9308o Directors: Cisco MDS 9506, 9509, 9513, 9530, 9560 [1]

? McData (now acquired and rebranded by Brocade):o Switches: 3232o Directors: 6064, 6140, 10000

? QLogic:o Switches: SANbox 5800, 5600, 5200, 3050, 1400o

Directors / Modular Chassis Switches: SANbox 9000

34461055-Storage-Basics

Documents

Transcript of 34461055-Storage-Basics