ST Unit 2

Infomation Storage and Management 2-1 Storage Networking Technologies
UNIT-II Mind sharing

file
Network:Storage Networking Technologies
Map
Components
Storage Implementation
2 Networking Technologies ,Benefits
Components
Syllabus DAS Benefits and Limitot

to SCSI : 1ypes of DAS, Model.
Direct-Áttached Storage and Introduction SCSI Command
Introduction to Parallel SCSI, Fiber Choa
Disk Drive Interfaces, SAN Evolution, SAN Components,
Storage Area Netsorks : Fiber Channel, Architecture, Zoning, Fiber Channel Login Tumo
NAS
FC_SAN
Connectivity, Fiber Channel Ports, Fiber Chanel Storage : Benefits of NAS, NAS File V/Componente
Attached
Fiber Channel Topologies. Network
NAS-Implementations, NAS File Sharing Protocols, NASVO
Operations
NAS, NASImplementations,
Contents
Technologies
21 Direct-Atached Storage Network
Storage
2.2 Disk Drive Interfaces
2.3 Introduction to Parallel SCSI
DAS
24 SCS Command Model
25 Storage Area Networks
2.6 Fibre Channel SCSI nnelhann
2.7 Components of FC-SAN
2.8 FC Connectivity
2.9 Fibre Channel Ports
2.10 Fitbre Channel Architecture
Types.
2.11 Fabric Service
disadvantages
and
Advantages
2.12 Zoning
2.13 Fibre Channel Topologies
2.14 Network AttachedStorage Topologies
2.15 Benefits of NAS
2.16 Network File Sharing model
Command
Architecture
-model
Communication
2.17 Components of NAS Architecture
2.18 NAS IO Operation Components
FC Ports
connectivity
2.19 NAS Implementation
2.20 NAS File Sharing Protocol
2.21 Factors Affecting NAS Performance
2.22 Application and Advantages of NAS
2.23 Two Marks Questions with Answers
TECHNICAL PUBLICATIONS- An up thrust for knowledge

(2- 0)
Storage Networking Technologles
2-2 Infomation Storage and Management 2-3 Storage Networking Technologiss
Information Storage and Management
2.1 Direct-Attached Storage

architecture ax3
(DAS) was the first networked storage :
" Direct Attached
Storage which the storage device Server Server Server Server Server
It is an architecture in
continues in wide use today.
directly attached to a server. built into a server and.
witL
storage device may include one or more drives, RAID array.
The a
adapter, may be configured as
an appropriate host bus internal hard drive in a laptonas Storage array
most familiar example of DAS is the
The
desktop PC. from DAS. With
protocol is used to access
data DAS storage
" Block level access server that contains the storage in
directly to the
architecture, clients must connect Fig. 2.1.1 External DAS
order to access the data.
down for maintenance, installation
of new hardware o 2.1.2 DAS Advantages and Disadvantages
" If the server should be becomes infected with vinusoe Advantages of DAS
or if it
application of operating system patches, data.
shared 1. DAS is ideal for localized file sharing in environments with a single server or a
clients would not be able to access
you need, all within one array. It's a few servers.
" DAS provides the flexibility and scalability requiring an efficient.
simple, yet powerful storage solution, ideal for applications 2. DAS devices can offer block-level access or file-level access
availability.
fully redundant, entry-level array to maintain 3. DAS also offers ease of management and administration.
and storage on external
" Examples of DAS include hard drives, optical disc drives, 4. DAS Can still be used locally to store less critical data.
drives. 5. Initial cost of DAS is lower than NAS.
disadvantage of DAS
" One advantage of DAS storage is its low initial cost. A key
storage is its limited scalability. A Host Bus Adaptor can only support a limited Disadvantages of DAS
number of drives. 1. DAS is limited in its scalability.
" Direct Attached Storage tends to be less expensive than NAS and SAN 2. DAS is limited to dedicated servers.
implementations, but lacks scalability and flexibility. 3. Unused resources cannot be reallocated easily.
2.1.1 Types of DAs 4. If the server the device is attached to is down for any reason, the data stored on
attached DAS devices is inaccessible.
" DAS are of two types : Internal DAS and External DAS
1. Internal DAS : Storage device is internally connected to the host. The connection 2.2 Disk Drive Interfaces
between host and storage device is serial or parallel bus. Physical bus is limited The interface protocols that connect to disk systems are Integrated Device
by distance. It also supports limited number of devices. Large storage space is Electronics (DE) /Advanced Technology Attachment (ATA) and Small Computer
also required. System Interface (SCSI).
2. External DAS : Server connects directly to the external storage device. Fiber
channel or SCSI is used for communication between the host and storage device. 2.2.1 IDEJATA
Fig. 2.1.1 shows external DAS.
" The Advanced Technology Attachment (ATA)-he official name for IDE
drives-standard is based on the original BM AT standard for hard disk drives.
TECHNICAL PUBLICATIONS - An up thrust for knowledge
Storage Networking Technolooi
2-4
Management
Jnformation Storage and
drive to a PC is typically called Infomation Storage and Management 2-5 Storage Networking Technologies
a hard disk
used to connect fact that the
interface
The primary
interface
Integrated Device
Electronics. IDE refers to the
separate board.
electronics ATA-3 " Eliminated single-word (&-bit) DMA transfer protocols
drive and is not a
controller is built into the controller, or interface, is
to " SMART (Self-Monitoring, Analysis, and Reporting Technology)
of the hard disk
The primary purpose the drive.
transmit and "Mandatory LBA support
from
receive data to and
meaning that 16 bits are " ATA security mode
" ATA is a 16-bit parallel interface, transmitted ATAJATAPI-4 " Also called Ultra-DMA/33
simultaneously down the interface. standarde " Integral ATAPI support
based on three bus
Four main types of IDE interfaces have been
" (SATA) " Optional 80-conductor, 40-pin cable
1. Serial AT Attachment ISA
Atachment (ATA) IDE based on 16-bit
2. Parallel AT 2.2.2 Serial ATA
ISA)
3. XT IDE (based on 8-bit
Micro Channel) " Serial ATA (SATA or Serial Advanced Technology Attachment) is a computer bus
4. MCA DE (based on 16-bit
today. The improved versions of interface for connecting Host Bus Adapters (HBA'S) to storage devices such as
" Out these four, only the ATA versions are used disks and optical drives.
ATA-2 and higher.
parallel ATA are referred to as
with Dine " Serial ATA is an enhancement to the ATA standard for for connecting storage to
" The ATAinterface connector is normally a40-pin header-type connector PC's. The ATA-7 specification introduced Serial ATA
spaced 0.1 inches apart and generally keyed to prevent the possibility of installine
it upside down. Serial ATA was designed to replace the older AT attachment standard (also
3. The ATA standard provides the option of operating on the AT bus in a known as EIDE). SATA offers several advantages over the older parallel ATA
daisy-chained configuration. Fig. 2.2.1 shows ATA cable. (PATA) interface, including smaller cables, lower cost, hot swapping, faster data
transfer, and more efficiency through an optional I/O queuing protocol.
" The original Serial ATA standard offered miniminal improvement over the existing
133 MBps bandwidth of the existing Parallel ATA standard
" Serial ATA standards are constantly evolving to meet higher bandwidth
requirements :
SATA Type Bandwidth Bus Speed Signal Rate

Fig. 2.2.1 ATA cable ATA-150 150 MBps 1500 MHz 15 Gbps
" The primary drive (drive 0) is called the master, and the secondary drive (drive 1)
SATA-300 300 MBps 3000 MHz 3.0 Gbps
is called the slave. ATA cable has 3
connectors.
SATA-600 600 MBps 6000 MHz 6.0 Gbps
Type Description
ATA- 1 " 40/44 pin " Serial ATA utilizes a point-to-point architecture. It removes the requirement for
connectors and cabling
"
bus arbitration and, no I/0 overhead is required for multiple device
Master/slave drive configuration communication.
" Signal timing for basic PIO and
DMA modes " SATA has improved signal control and clocking as well as a hot-swap interface,
ATA-2 " Faster PIO and DMA
s*******
transfer modes which allows for higher frequencies with a lower risk of data loss
" Defined standard
up to 8.4G
CHS/LBA translation methods for support drives
TECHNICAL PUBLICATIONS- An up thrust for
knowledge TECHNICAL PUBLICATIONS - An up thrust for knowiedge
Storage Networking Technologie
2-6
Management
InformationStorageand Information Storage and Management 2-7 Storage Natworking Technologies
Introduction to Parallel SCSI
2.3 2. SCSI devices are "smart" devices with their own control circuitry. They can
System Interface (SCSI).
SCSIis also referred to
as aSmall Computer "disconnect" themselves from the host adapter to process tasks on their own, thus
" Parallel interface
a computer bus and a set freeing up the bus for other transmissions.
Small Computer System Interface (SCSI) is transfer between of
" connectors and data computers and data 3. The bus can handle simultaneous reads and writes.
standards for physically
storage devices. Evolution and SCSI Interface:
" The SCSI standards defines protocols, commands, and electrical and optical
" SCSI-1 : It is basic SCSI version and approved by ANSI. It define cable length,
interfaces.
as 7 or 15 total devices. The SCSI
command and transfer modes, characteristics etc. It supported only by single
SCSI is a bus that supports as many
called the host adapter, functions as the
gateway between the SCSI bus and controler, ended transmission.
" SCSI-2 :It focus on improving performance and additional features of the SCSI-1
PC system bus. interface.
The SCSI bus does nos .,
. Each device on the bus has a controller built in.
directly with devices such as hard disks; instead, it talks to the controller that is SCSI Type Speed Bus width
built into the drive. SCSI-1 5 MB/S 8
" SCSI is normally implemented in host adapters, which provides a shared bus t Fast SCSI 10 MB/s
which peripheral devices attach. Fast Wide SCSI 20 MB/s 16
. SCSI1accommodated a maximum transfer speed of 5 MBps, at 5 Mhz on an 8-bit Ultra SCSI 20 MB/s
bus which means that it was a 'narrow' implementation
Wide Utra SCSI 40 MB/s 16
" In theSCSI cable, each bit of a data byte is given its own wire. A SCSI cable may
Ultra2 SCSI 40 MB/s
have 8 (narrOw SCSI) or 16 (wide SCSI) data lines, plus control lines to manage
device selection and data transmission. Wide Ultra2 SCSI 80 MB/s 16
" Communication on the SCSI bus Request Ultra 160 SCSI 160 MB/s 16
Client machine Server machine
takes place between a device that Ultra 320 SCSI 320 MB/s 16
initiates the transfer and (Initiator) Response (Target)
destination device. Serial Attached SCSI 3 Gb/s N/A
Fig. 2.3.1
shows SCSI Communication. Fig. 2.3.1
" At any time, communication is performed only between two devices, one initiator 2.3.1 SCSI-3 Architecture
that selects and controls the target device that
performs the requested operation. " Fig. 2.3.2 shows SCSI-3 architecture and it was approved and published as the
Clients called Initiators issue requests and Severs standard X.3.270-1996 by the ANSI.
called
initiator requests. SCSI commands sent in units called Targets respond
Blocks. Command Descriptor Blocks (CDB) is the data Command Descriptor " This architecture helps developers, hardware designers, and users to understand
the command parameters that are to structure used to contain and effectively utilize SCSI.
be handed by an initiator to a " SCSI architectural model consists of three components:
" Some of the target.
features of the SCSI interface are list
1. The original SCSI standard below : 1. SCSI-3 command protocol : It contains primary commands which are common to
new standards support
supports up to 7 devices on a single host all devices as well as device-specific commands that are unique to a given class of
lengths of up to 12 meters. high-speed operation with up to 16 devicesadapter, bu!
and bus
devices.
2. Transport layer protocols : These are a standard set of rules by which devices
communicate and share informnation.
TECHNICAL PUBLICATIONS - An up thrust for TECHNICAL PUBLICATIONS - An up thrust for knowledge

knowledge
Storagê NetwoRIng |echnologieg
2-10
Management
Information Storage and Information Storage and Management 2-11 Storage Networking Technologies
Communication Model
2.3.2 SCSI contains three layers. The first byte of a SCSI CDB shall contain an
communication model. It operation code identifying the
" Fig. 2.3.4 shows
SCSI operation being requested by the CDB
Target device Each CDB can be a total of 6, 10, 12, or 16 bytes, but later
Initiator device versions of the SCSI
Application
SCIS Application layer standard also allow for variable-length CDBs.
SCIS application
application protocol Command Descriptor Block (CDB) : The structure used to communicate
Transport Transport layer
from an application client to a device server. A CDB nay have a fixedcommands
length of
Transport Target protocol up to 16 bytes or a variable length of between 12 and 260 bytes.
protocol
protocol
" Fig. 2.4.1 shows CDB structure.
Interconnect Interconnect layer
Interconnect Interconnect
services Bit
services
Byte 7 6 5 4 3 2
Operation code
model
Fig. 2.3.4 SCSI communication 1
Reserved Command parameters
2 to 8 Command parameters
" Three layers are as follows : 9
It iniiak Control byte
1. SCSI application layer (SAL) : It contains client and server applications.
and process SCSI I/O operations using a SCSI application protocol. Fig. 2.4.1 CDB structure
2. SCSI Transport Protocol Layer (STPL) : It is used for communication between an The CDB consists of a
one-byte operation code followed by some
initiator and targets. It contains the services and protocols for communication. command-specific paramneters. The parameters need not be a full byte long, and
3. Interconnect layer : This layer facilitates data transfer between the initiator and the parameter length varies from one command to another
targets. The interconnect layer is also known as the service delivery subsystem The CDB defines the operation to be performed by the
device server.
and comprises the services, signalling mechanisms, and inter-connects for data For all commands, if the logical unit detects an invalid
transfer. parameter in the CDB, then
the logical unit shall complete the command without
information.
altering the media
2.3.3 Difference between SCSI and IDE
All CDBs shall have an OPERATION CODE as the first byte
SCSI IDE Operation code:
Data transfer speed is 320 MB/S Data transfer speed is 100, 133 and 150 MB/s Operation code consists of command and group code. Fig. 2.4.2 shows operation
code.
It supports 16 devices It supports 2 devices
Performance is high Performance is moderate to low Bit
Itsupports internal and external connectivity 7 6 5 4 3
It supports internal connectivity 2 1
Ease of Configuration is high
Ease of Configuration is low Group code Command code
2.4 SCSI Comnmand Model Fig. 2.4.2 Operation code
" In SCSI standards for
transferring data between computers and
often computer storage,
The three-bit GROUP CODE field provides for eight groups of
command codes.
(CDB). commands are sent in a Commandperipheral deyD The five-bit COMMAND CODE field provides for thirty-two command
codes in
Descriptor each group. A total of 256 possible operation codes exist.
TECHNICAL PUBLICATIONS - An up
thrust for knowledge TECHNICAL PUBLICATIONS- An up thrust for knowledge
2-12
Storage NetworKing Technologia.
Management
and Information Storage and Management
InformationStorage 2-13 Storage Networking Technologies
below:
commandsare listed
SCSI
" Common Remarks SANs are primarily used to enhance storage devices, such as disk arrays, tape
Command
libraries, and accessible to servers so that the devices appear like locally attached
Read data fromn a
device devices to the operating system.
READ
Writes data to a
device " SAN systems provide a simple, untyped, fixed-size (block), memory-like interface
WRITE
Returns basic information for manipulating nonvolatile magnetic media. SAN is used when multiple servers
INQUIRY
need access to the same data (example : databases).
numbers
Lists the logical unit A SAN is a high-speed network of storage devices, separate from the local area
REPORTS LUNS
whether it is ready for data transfer
TEST UNIT READY Oueries the device to check network that also connects those storage devices with servers. It provides
from log pages block-level storage that can be accessed by the applications running on any and all
LOG SENSE
Returns current information servers connected to the SAN. SAN storage devices can include tape ibraries and
pages of a SCSI target device
LOG SELECT Used to modify data in the log more
commonly, disk-based devices, like RAID hardware.
mode pages
MODE SENSE Returns current device parameters from A storage area network is designed for managing very large amounts of network
page storage. For large organizations with many servers that need access to the same
MODE SELECT Sets device parameters on a mode
data, SAN offers better performance and flexibility than DAS or NAS.
" Storage area network is defined as a set of interconnected devices and servers that
2.5 Storage Area Networks are connected to a common communication and data transfer infrastructure such
" SAN stands for storage area network. A
storage area network is a dedicated as fiber channel. The purpose of the SAN is to allow multiple servers access to a
storage. ig. 2.5.1
network that provides access to consolidated, block level data pool of storage in which any server can potentially access any storage unit.
shows storage area network. A SAN is a network designed to transfer data from servers to targets and it is
alternative to directly attached target architecture, or to a DAS architecture, where
Client LAN /WAN the storage is connected to the servers on general purpose networks.
A SAN can be considered as an extended and shared storage bus within a data
center, consisting of various storage devices and specific interfaces (eg.fibre
channel ) rather than the Ethernet. SANs provide high-bandwidth block storage
access over long distance via extended fiber channel links.
Server
SAN features
1. Network resource used exclusively for the storage.

2. SCSI protocol for end to end communication: minimum impact for DAS to SAN
Storage migration.
area
network 3 Access to data through logical blocks and not to fle.
4. Compatibility with an high number of nodes.
Disks
5. High reliability and ability to react to failures.
6. Performance allows concurrent access of disk or tape arrays by two or more
servers at high speeds.
Fig. 2.5.1 Storage Area Network 7. Availability have disaster tolerance built in, because data can be mirrored using a
SAN up to 10 km or 6.2 miles away.
TECHNICAL PUBLICATIONS - An up thrust for knowledge TECHNICAL PUBLICATIONS- An up thrst for knowiedge
Storage Networking Technologi
2-14
Management Information Storage and Management 2-15
Information Storage and Storage Networking Technologies
technologies. This allowe
LAN/WAN, it can use a variety of 2.6.1 Fibre Channel Architecture
8. Scalability like a data replication be
operations, file migration and
relocation of backup data, Fibre channel transfers digital data between sources and users of
systems. inforrmation. This
digital data represents different types of information like programs, files, graphics,
2.6 Fibre Channel videos and sound. Each having its own structure, protocol, connectivity, measures
chan of performance and reliability requirernents. Fig. 2.6.1 shows
transmission technology used sa peripheral network connection
" Fibre channel is a high-speed serial, data communicat example.
or networkbackbone. It is a 100 MB/sec, full-duplex,
technology.
protocols like internet protocol and SCSI Client machine Client machine Client mnachine
" FC supports several common transport
operates over copper and fiber optic cables at distances of up to 10 kilometers.
Fabric
which is based on a series of ANC
. Fibre channel is a multi-layered network, Dort
standards that define characteristics and functions for mnoving data acrOss A
network. These standards incude definitions of physical interfaces. Example. Storage devices
include : Cabling, distances, and signaling; data encoding and link controls; dab Printer
delivery in terms of frames; flow control and classes of service; common services Server
and protocol interfaces.
" Fibre charnel architecture includes a significant amount of hardware processing to Node port
deliver high performance.
" Fibre channel uses a serial data transport scheme, similar to other comnputer
networks, which stream packets (frames) of bits, one behind the other, in a single Fig. 2.6.1 Network connection
data line to achieve high data rates.
" Fibre channel is designed to transport many protocols, such as FDDI, serial HIPPI,
FC offers a layer 2 transport services, and uses layer 3 services to provide it, and
to map upper layer protocols like SCSI, HIPPI, IP or others.
SCSI,IPL.
1. FCO defines the physical interface (data rates, connectors, media, distances, power
" Why use Fibre Channel for SAN ? and so on). This also includes electrical and optical characteristics, transmission rates,
" Fibre charnel is the leading technology for and other physical components of the standard.
implementing SANs today because it
meeting the requirements of today's applications. FC is fast, 2. FC1 deals with encoding/decoding of bits on the physical interface and low level
speed of up to 1 Gbps with 2, 4 and 10 Gbps in the works. currently supports
it
signaling. Every 8 bits of data are encoded into a 10 bit transmission character.
It allows storage devices to be accessed by all 3. FC2 defines framing and classes of services. FC-2 is called the signaling protocol
servers on the SAN thus
reliability. It supports several different topologies. The topologies haveimproving level. It is responsible to break the data to be transmitted into frame size, and
levels of cost to capabilities tradeoffs, thus varying reassemble the frames after transport. It specifies the framing rules of the data to be sent
small, reliable setup and scale up as needed. allowing organization to start with a from one port to another, ways for controlling the three service classes, and controlling
In general fibre channel provides speed, the sequence in which the data is transferred.
reliability, distance, and
at a low cost. It guarantees
any protocol or bandwidth and delivery while being connectivity; al
able to handle
4. FC3 provides common services like multicasting or possibly encryption.
topology. 5. FC4 maps upper layer protocols onto fiber channel. This layer defines the
application interfaces that can be run over fibre channel. Both network and channel
protocols can concurrently run over the same physical interface providing seamless
integration of starndards
TECHNICAL PUBLICATIONS- An up thrust for TECHNICAL PUBLICATIONS- An up thrust for knowledge
knowledge
2-16 Storage Networking Technolooio.
Information Storage and Menagement Information Storage and Management 2-17 Storage Networking Technologies
FC-SAN
2.7 Components of Outer
storage and network infrastructure. jacket Strength
component of SAN is server, node ports, cabling, These member
" Basic key elements are
componernts consists of some management software.
storage array
interconnecting device and SAN Cladding glass
1. Node Port library. These all devices
hosts, storage and tape Core glass
" Fibre channel consists of destination Dork
one or more source or
called as nodes. Every node is
information transfer. For communicating with other node, physical interface : Inner Fiber
ports
required for every node port. These and Transmit data
Coating
of an HBA Port 0
are integral components Receive data Fig. 2.7.2 Single mode
storage front adapters.
data
" Node port operates in full duplex
Port 1 Cladding glass
transmit
transmission mode. It also uses
link and a receive link for communication Port 2
with other device. Fig. 2.7.1 shows the
nodes, ports and links. Fig. 2.7.1 Node Core glass
" A node may be a workstation, disk drive
Fig. 2.7.3 Multimode cable
or disk array, a medical scanning instrument, etc. Each node can have more than
one port. A port can be an adapter including a part of software in the host: The index of refraction between the core and cladding of single mode changes less
sometimes it is difficult to say exactly where a port ends and the node begins. than it does for multimode fiber, creating little dispersion.
Some parts of software may be shared by several ports. Fig. 2.7.4 shows standard connector.
2. Cabling A fiber-optic cable connector that
SC
" Fiber optic cable is used for implementation. For back end connectivity and shorter uses push-pull latching
distance copper cable can be used. The mode of a fiber refers to the number of mechanism similar to common
paths for the light rays within the cable. Fiber optical cable are of two types audio and video cables. For
a. Single mode fibers are capable of carrying only one signal of a specific bi-directional transmission, two fiber La
wavelength. Single mode fiber allows propagation of light ray by only one path. cables and two SC connectors (Dual
Intermodal dispersion does not present. Fig. 2.7.2 shows single model fiber. Due Fig. 2.7.4 SC Connector
SC) are used.
to little signal dispersion, single mode fiber is widely used for long-distance
SC connectors are used with single-mode and multimode fiber-optic cables. They
transmission. Optical media converters are used to do the fiber to fiber conversion
to connect two different transmission modes, making single mode offer low cost, simplicity and durability. LC connectors are used with single-mode
fiber efficiernt in and multimode fiber-optic cables. Two simplex SC connectors can be clipped
long-distance transmission. Fig. 2.7.2 shows single mode. together by a reusable duplex holding clip to create a duplex SC connector.
b. Multimode fiber allows multiple paths for accurate
weakens the signal, reducing the length and speed
light ray to propagate. This strateg) " The LCconnectors are constructed with a plastic housing and provide for
Intermodal dispersion exists. Multimode fiber transmits the data signal can travel. alignment via their ceramic ferrules. LC connectors have a locking tab. LC
multimode paths for its larger core. Those multimode fibers signals throug connectors are rated for 500 mating cycles.
distortion at the receiving end in long-distance can cause signa
unclear and incomplete. Fig. 2.7.3 shows transmission, making signa
multimode cable.
TECHNICAL PUBLICATIONS- An up thrust for knowiedge
knowledge
Storage Networking Technologji
2-18
Management Information Storage and Management 2-19 Storage Networking Technologies
and
rmationStorage
Interconnectdevice interconnectingdevices, " Using specialized algorithms, several drives are grouped to
provide common
directors as FC-AL. levels, define
SAN uses hub, switchand communication device used in Star topology pooled storage. These RAID algorithms, commonly known as RAID
" physical connected with hub,
is
a. Hub : Hub for connecting nodes.
Device
routing for a Fibre Channel
h shares the characteristics of the particular grouping
and replication features
hub device
bandwidth. Ahub
provides centralized
devices directly
together, each device is connected arbitrale " Most storage arrays provide additional data protection
such as snapshots, internal copies and remote mirroring
loop. Instead of
connecting connection from device to device. Fig. 275
hubcompletes the SAN management software
2ort on a hub. The connections, storage elements and
hub with device. " SAN management software organizes the
s system. SAN management software does this
Hub computer systems in a SAN storage
so that data transfer is secure and robust.
planning and monitoring of storage
" SAN management software can also help with mid-sized businesses
systems. The software can be provided for large enterprises,
commercial product or as
and small companies. The software may be offered as a
solution.
Disk Tape an open-source or free
Cient Server aray library device discovery, allocation, RAID
machine
" The management applications (zoning, environment of distributed systems.
Fig. 2.7.5 hub subsystems and others) manage the complex
and improve the quality of
These applications can significantly reduce the cost
hub port has a port bypass cirguit
. The mn advantage of a hub is that each responding, it closes the PBC
enterprise information services.
PBO Ea port detects that a device is absent or not connection
allowing the loop to remain intact. This prevents a
failing device or 2.8 FC Connectivity
from bringing down the entire loop. area networks today is Fibre Channel.
the The primary technology used in storage
b. Switch : Switch is more intelligent than hub and they do not share Fibre Channel Topologies defines three configurations :
bandwidth SAN Switches can connect both servers and storage devices and thus
a. Point-to-point
provide the connection points for the fabric of the SAN. They are the medium
through which servers and storage devices communicate to each other. As such, b. Fibre Channel Arbitrated Loop (FC-AL)
they have physical ports used to interconnect servers, storage devices and other c. Switched Fibre Channel Fabrics (FC-SW)
switches. FC terminology includes the concept of port typing. For smaller SANS,
the standard SAN switches are called modular switches and can typically suppot 2.8.1 Point-to-Point
8 or 16 ports. Sometimes modular switches are interconnected to create a Point-to-point fibre channel is a simple
fault-tolerant fabric. For larger SAN fabrics, director-class switches provide a way to connect two and only two
larger port capacity (64 to 128 ports per switch) and built-in fault tolerance. devices directly together, as shown in
" The fiber channel standards specify electrical and optical transmission media, as channel
Fig. 2.8.1 below. It is the fibre Host Storage
well as conventions for signaling and transmission/functional protocols. Optcal equivalent of Direct Attached Storage configuration
Fig. 2.8.1 Point to point
medium (with SC, LC, and MT-RJ connectors) supports reliable signaling ove (DAS). other device's
long distances. connectionis directly linked by a Fibre to the
" One device's "ransmit"
Storage arrays "receive" connection, and vice versa.
Storage arrays rarely provide hosts direct access to configuration.
storage array uses RAID individual drives. Instead, Point-to-point is not a scalable enterprise
technology to RAID uses offers limited connectivity.
group a set of drives.
independent drives to provide capacity, performance " Point to point configuration
and redundancy.
thrust for knowledge
TECHNICAL - An
Storage Networking Technologies
2-18
end
Management Information Storage and Management 2-19 Storage Networking Technologies
InformationStorage
interconnecting devices.
3. Interconnect device
directors as Using specialized algorithms, several drives are grouped to provide comnon
usedin FC-AL. Star
switch and
SAN uses hub, physical communication device connected with hub, shares topology usey pooled storage. These RAID algorithms, commonly known as RAID levels, define
the characteristics of the particular grouping
Hub : Hub is connecting nodes.
Device
Channel the
routing for a Fibre
a.
hub device for

bandwidth. A hub
provides centralized
together, each device is connected arbitrattoeda Most storage arrays provide additional data protection and replication features
such as snapshots, internal copies and remote mirroring.
connecting devices directly from device to device. Fig.
loop. Instead of hub completes the connection
The
275 SAN management software
port on a hub. device. " SAN management software organizes the connections, storage elemernts and
shows hub with Hub computer systems in a SAN storage system. SAN management software does this
so that data transfer is secure and robust.
SAN management software can also help with planning and monitoring of storage
systems. The software can be provided for large enterprises, mid-sized businesses
and small companies. The software may be offered as a commercial product or as
Disk Tape an open-source or free solution.
Client Server array library
Cient
machine
machine " The management applications (zoning, device discovery, allocation, RAID
Fig. 2.7.5 hub subsystems and others) manage the complex environment of distributed systems.
a port bypass cirei These applications can significantly reduce the cost and improve the quality of
main advantage of a hub is that each hub port has the PRC enterprise information services.
" The
absent or not responding, it closes
(PBC), If a port detects that a device is prevents a failing device or connectiom
allowing he loop to remain intact. This 2.8 FC Connectivity
from bringing down the entire loop. share the " The primary technology used in storage area networks today is Fibre Channel.
intelligent than hub and they do not
b. Switch : Switch is more servers and storage devices and thus Fibre Channel Topologies defines three configurations :
bandwidth. SAN Switches can connect both a. Point-to-point
SAN. They are the medium
provide the connection points for the fabric of the
communicate to each other. As such. b. Fibre Channel Arbitrated Loop (FC-AL)
through which servers and storage devices
storage devices and other c.Switched Fibre Channel Fabrics (FC-SW)
they have physical ports used to interconnect servers,
SANS
switches. FC terminology includes the concept of port typing. For smaller 2.8.1 Point-to-Point
the standard SAN switches are called modular switches and can typically support
8or 16 ports. Sometimes modular switches are interconnected to
create a Point-to-point fibre channel is a simple
fault-tolerant fabric. For larger SAN fabrics, director-class switches provide a way to connect two and only two
larger port capacity (64 to 128 ports per switch) and built-in fault tolerance. devices directly together, as shown in
" The fiber channel standards specify electrical and optical transmission media, a Fig. 2.8.1 below. It is the fibre channel Hast Storage
well as conventions for signaling and transmission,/functional protocols. Opucu equivalent of Direct Attached Storage Fig. 2.8.1 Point to point configuration
medium (with SC, LC, and MI-RJ connectors) supports reliable signaling ove (DAS).
other device's
long distarnces. " One device's "transmit" connectionis directly linked by a Fibre to the
"receive" connection, and vice versa.
Storage arrays
" Storage arrays rarely provide hosts direct access to individual drives. Instead, the " Point-to-point is not a scalable enterprise configuration.
storage array uses RAID technology to group a set of drives. RAID uses
" Point to point configuration offers limited connectivity.
independent drives to provide capacity, performance and redundancy.
TECHNICAL PUBLICATIONS An up thrust for knoedodee

2-20 Information Storage and Management 2-21
Information Storage andManagement Storage Networking Technologies
Arbitrated Loop . In a switched fibre channel fabric,

2.8.2 Filbre Channel
Host A Host B Host C Host D
devices are connected in
that are
hosts and devices many-to-many topology using fibre
" It is a set of 2.8.2
connected into a single
loop. Fig. channel switches.
arbitrated loop. Devices are
shows the device's When a host or device communicates
loop, each Host A Host B
connected in a dosed with another host or device, the source
to the next
transmit port linked directly and target setup a point-to-point Switches fiber channel fabric
device's receive port. Device E Device C connection between then and
connect up to
" It is a cost-effective way to communicate directly with each other.
126 devices and hosts into a single
. In a fibre channel switched fabric, the DeviceDevicelDevicelDevicelDevice
network. Device D E F
media is not shared. Any device can
" Devices on the loop share
the media; Fig. 2.8.2 Fibre channel arbitrated loon
the communicate with any other device Fig. 2.8.3 Fibre channel switched fabric
each device is connected in series to and communication occurs at full bus
next device in the loop and so on around speed irrespective of other devices and hosts communicating
pass through all
the loop. Any packet traveling from one device to another must
intermediate devices. " Fig. 2.8.3 shows switched fibre channel fabrics.
Using arbitration protocol, a single connection is established between a sender and Each port in a fabric has a unique 24-bit fibre channel address for communication.
a receiver, and a data frame is transferred around the loop. When fhe Advantages :
Communication comes to an end betweern the two connected ports, the loon 1. Easy to deploy.
becomes available for arbitration and a new connection may be established.
2. Supports 16 million hosts and devices.
" No sophisticated hardware is required to set up arbitrated loop. The only
necessary components are fibre channel devices or bridges, host adapters and 3. Communicate at full wire-speed, no shared media.
driver software for host computers, and cabling. Fibre channel hubs and dual 4. Switches provide fault isolation and re-routing.
loops, however, boost reliability and availability for arbitrated loop topologies. Disadvantages :
Advantages : 1. Difficult for vendors to develop.
1. Low cost.
2. Interoperability issues between components from different vendors.
2. Loops are easily expanded and combined with up to 126 hosts and devices. 3. Switches can be expensive.
3. Easy for vendors to develop.
2.9 Fibre Channel Ports
Disadvantages :
1. " Each node must have at least one port (hardware interface) to connect the node to
Difficult to deploy. the FC topology. This node port is referred to as an N_Port.
2. Maximum 126 devices.
Other port types include :
3. Devices share media
thus lower overall 1. N_Port : It is also called as node port. Node ports used to connect devices to
bandwidth.
2.8.3 Switched Fibre Channel switched fabric or point to point configurations.
Fabrics 2. NL_ Port : This port is also known as node loop port. Loop ports are used in a
" It is also reffered as
fabric connect. A fabric is a Fibre Channel Arbitrated Loop (FC-AL) to build networks without FC switches.
communicate with one another in a network. logical space in which all nodes
TECHNICAL PUBLICATIONS- An up thrust for knowiedge
knowledge
Storage NetworkingTechnologia,
2-22
Managenment Information Storage and Management 2- 23 Storage Networking Technologies
and switches and build a
InformationStorage
expansion port is
usedto
interconnect
switche FC- 4 Upper-layer protocols (PDU): SCSI,IP
3. E_Port
: An fabric.
fabric. located in the
shows various FC ports FC -3 Not implemented
Fig. 2.9.1
FC -2 Framing, flow control, zoning, ordered sets
N_Port
FC-1 Signalling, encode/decode including 8b/10b
FC switched
NPort FPort/
fabric
Port
E Port FC -0 Lasers, cables, connectors, data rates(x Gbps)
E Port Fig. 2.10.1 Fibre channel protocol stack
3. FC-2 is called the signaling protocol level. It is responsible to break the data to be
F Port transmitted into frame size, and reassemble the frames after transport. It specifies
NPort
the framing rules of the data to be sent.
FL Port
4. FC3 Layer : FC-3 provides the Fibre Channel Generic Services (FC-GS) that is
required for fabric management. Specifications exist here but are rarely
NL Port NL Port impiemented.
5. FC-4 : This layer defines the application interfaces that can be run over fibre
NL Port NL Port channel. Both network and channel protocols can concurrently run over the same
Fig. 2.9.1 Port in fabric
physical interface providing seamless integration of standards.
4. F Port : A fabric port is used to connect an N_Port to a switch that is not

2.10.1 Fibre Channel Addressing
loop-capable. " Addresses are assigned
dynamically to port when it 24- Bit Address Space
5. FL_Port : Afabric loop port is used to connect NL Ports to a switch in a loop
configuration. logs on the fabric. FC address
Domain ID Area ID Port ID
format is changed according to
6. G_Port :A generic port is a port that has not assumed a role in the fabric. node port in the fabric.
(One byte) (One byte) (One byte)
2.10 Fibre Channel Architecture " Node ports are N_port and 1-239 0-255 0-255
" Fig. 2.10.1 shows fibre channel protocol stack. NL_port in a public loop and
1. FC-O Layer : It defines the physical NL_port in a private loop. |ldentifies the ldentifies the Identifies the
transmission medium and specifies which Fig. 2.10.2 shows 24 bit FC
switch number switch port N_port
physical signals are used to trarnsmit the bits 0 and 1.
address of N_port. Fig. 2.10.2 Bit FC address ofN_port
2. FC-1 defines all transmission
protocols
rules, special characters and error control.including8 serial encoding and decodiig The 24-bit field is composed of three bytes :
10 bit Every bits of data are erncoded nto a
transmission character. 1. First byte (Domain ID) : This is the address of the FC switch or director. It is
unique to the fabric. There are theoretically (256 - # reserved addresses) = 239
switches possible in a SAN fabric.
TECHNICAL PUBLICATIONS- An up thrust for TECHNICAL PUBLICATIONS - An up thrust for knowledge

knowledge
Storag
Information Storage and

Management 2-24
|echnologios
used to identify the Information Storage and Management
Ntindiypeviduapon
of the address is 2-25
Second byte (Area ID) : This part 3rd byte. Identifies
2.
The Area ID is very similar to the Start of Frame Frame Header
switch ports. Data Field CRC End of Frarne
a switch. [4 Bytes] (24 Bytes]
that is connected to [O to 2112 Bytes] [4 Bytes]
identify a single FC object on the [4 Bytes]
: This address is used to Fig. 2.10.3 Frame format of FC
3. Third byte (Port ID) addresses for identifying attached
N_Ports. 1. Start of Frame
frame
: SOF acts as a delimiter. It is one type of
SAN fabric. It provides 256 simple calculation flag and indicates
" To arrive at the
number of available addresses
is a based on: whether the frarme is the first frame ina sequence of
frame.
Domainx Area x Ports 2. Frame Header: It contains addressing information of the frame. Header
addresses available
256 x 256 = 15,663,104 information can includes source ID, destination ID, sequence number etc. The
This means that there are 239 x frame header is used to control link applications, control device protocol
and detect missing or out of order franmes.
transfers,
2.10.1.1 World Wide Names
addresses that identify them in an Do
Nodes and ports have unique 64-bit 3. The 4 bytes Cyclic Redundancy Check (CRC) precedes the EOF delimiter. CRC is
assigned by the manufacturer with
topology. These addresses are committee.
used to verify the integrity of the frame header and payload fields. This helps
vendor-specific portion defined by the IEEE standards detect errors in a frame.
. WWNs are similar to the Media Access Control (MAC)

addresses used in 4. Data Field : It contains user data.
networking. WWNs are burned into the hardware or assigned through software 5. End of Frame :End-of-Frame (EOF) delimiters signal the end of a frame.
ID, which is called
" Each participant in the fibre channel environment has a unique
the World Wide Name (WWN). This WWN is a 64-bit address, and if two WWN 2.10.3 Structure and Organization of FC Data
addresses are put into the frame header, this leaves 16 bytes of data just for " Data transport is communication between two person and frame represents word,
identifying destination and source address. So 64-bit addresses can impact routing a sequence represents a sentence, and an exchange represents a conversation.
performance.
" Fig. 2.10.4 shows exchange, sequence and frame relation.
" World wide names are of two types : World Wide Node Name (WWNN) and
World Wide Port Name (WWPN) Exchange
" AWorld Wide Node Name (WWNN) is a globally unique 64-bit identifier that is Sequence Sequence Sequence
assigned to each Fibre Channel node or device. For servers and hosts, WWNN is Frame Frame Frame Frame Frame Frame
unique for each HBA (host bus adapter), and in a case of a server with two HBAS,
they have two WWNNS Fig. 2.10.4 Exchange, sequenca and frame relation
" A World Wide Port Number (WWPN) is unique identifier for each FC port of 1. Exchange operation : It takes place between two N_ports. It identifes and
any
Fibre Channel device. For server, we have a WWPN for each port of the HBA. For manages a set of information unit. Unit maps to a sequence. Exchange sequence
a SAN Switch, the WWPN is available for each port in the may be unidirectional or bidirectional.
chassis. For Storage,
each host port has an individual WWPN.
2. Sequence : It is a contiguous set of frames that are sent from one port to another.
" The WWNN is used to identify the switch or
the host, while the WWPN is used Asequence corresponds to an infomation unit, as defined by the ULP.
to identify individual ports on the switch or the
host. 3. Frame : A frame is the fundamental unit of data transfer at Layer 2. Each frame
2.10.2 FC Frame can contain up to 2,112 bytes of payload.
" The frames contain the information to be 2.10.4 Flow Control
source and destination ports and link transmitted (Payload), the address of the
"
control information. " The framing protocol must be concerned about how many frames one source
Fig, 2.10.3 shows frame format of
FC frame. It consists of
five fields. N_Port can send to another with-out overflowing the buffers in the receiving
NPort. To address this problem, there are two types of credit.
TECHNICAL PUBLICATIONS - An up thrust for
knowledge TECHNICAL PUBLICATIONS - An up thrust for know/edge
Storage Networking Technologia. Information Storage and Management 2-27
2-26 Storage Networking Technologies
Management
Informetion Storage and
mechanism : Buffer to buffer credit and End Advantages and disadvantages of Class-2 and Class-3 service
uses two flow control
technology
" FC Advantages :
to end credit. credit is associated with only
(BB_Credit): This type of receiver (F Port .The transmission reSources can be shared
1. Buffer-to-buffer
credit
transmitter to the next or N_Port), anong many users on a link.
exiting the
the immediate fiber primitive signal on a link. Due to frarne switching, the links are
occupied only for the
by the R_RDY
ransmission, making the path available for frarnes generated byduration
is managed of frame
This credit negotiated between a so
End-to-end credit (EE_Credit) : This type of credit is is managed using increasing both the link and Fabric bandwidth utilization.
other sources and
2. destination NPort. This credit
N Port and a back to the so
from the destination
acknowledgment (ACK) frame sent 3.
Erames in the same intormation block can be transmitted
sirnultaneously on
is not used in Class different paths in the Fabric.
NPort. This type of credit
Disadvantages :
2.10.5 Class of Service
connection-oriented and based on. 1. Out-of-order frame delivery in the generic Fabric case.
channel class-1 service is
a. Class-1 service : Fibre performed between connections
circuit-switched mode of operation, where switching is bandwidth guaranteed between 2. In transactions between many devices and one device, improper buffer allocation
Class-1 provides a dedicated connection with the full in the devices may result in long queuing delays within the Fabric.
two communicating devices.
3. Due to the lack of a dedicated connection, the transfer of very large blocks of raw
Advantages :
information can cause long fabric delays in a congested Fabric
1. Guaranteed in-order delivery of frames. 2.10.5.1 Comparison between Classes of Service
connection can interact.
2. Users and applications on a dedicated
satisfied that all information Parameters Class-1 service Class-2 service Class-3 service
3. The connection is not terminated until both parties are
has been correctly transmitted. Connectionless
Connection type Connection-oriented Cornectionless
Disadvantages : End to end credit and
Flow control End to end credit B-to-B credit
1. Possibility of bandwidth starvation to the devices wanting to communicate. B-to-B credit
2. The time to set up and terminate a connection may take much longer than the No guarantee of
amount of time required to transfer an information block. Delivery of frame Order delivery No guarantee of order
order
3. During connections,bandwidth is wasted when the links are idle while devices are
performing other tasks. Frame acknowledgement Acknowledged Acknowledged Not acknowledged
b. Class-2 service : This service is connectionless and based on a frame-switched Moderate High
Utilization of bandwidth Poor
mode of operation where switching is performed for each frame. In Class-2, frames can
be multiplexed between a source device and one or more destination devices. Support for multiplexing No Yes Yes
c. Class-3 service : It is also a connectionless service based on frame switching Supported

Unlike Class-2, it is a datagram service, which does not provide confirmation of delivery Broadcast and multicast Not supported Not supported
or non-delivery of the frames at the destination devices. The confirmation of delivery s Support
delegated to the upper-level protocol, such as TCP, which manages its own end-to-ena
flow control and data integrity. Only link-level flow control exists between the Fabric
and each connected device. There is no acknowledgment of a frame or
frame has been rejected or busied by the Fabric or by the notification that a
destination device. thrust for knowiedge
2-28
Infomation Storage and Management lnformation Storage and Management 2-29
2.11 Fabric Service imnlement compartmentalization of data for security purposes. Each device in a
SAN may be placed into multiple zones.
" Fabric services are of
following types :
Controler 4. Management Serv Ahen a device logs onto a fabric, it is
Server 2. Name Server 3. Fabric registered with the
1. Fabric Login
assigns a fabric address to a fabric
node, which allows i . shows the zoning. Fibre channel zoning allows user to name server. Fig. 2.12.1
" The fabric login server other nodes in the fabric. Thbe f partition the fibre channel
switch or fabric into one or more zones.
communicate with services on the
three 3-byte nodes.
address is a 24-bit address containing switches :
implement a concept known as the name server. All
Fabric switches updated, and are therefore aware of all
server Server 2
the fabric keep the simnple name Server 3
devices in the name server.
provides State Change Notifications (SCNA
" Fabric controller : The fabric controller
fabric topology occurs.
to registered nodes when a change in the
management server provides a single point for
" Management server : The configure.
managing the fabric. This is the only service that users can
Fibre channe! storage
2.11.1 Switched Fabric Login Type Server 1 area network Server 4
Fabric login
" Login types of three types : Port login, Process login and
" Port login : Port login is also known as PLOGI. It is used to establish a session
between two N_Ports and is necessary before any upper level commands or Storage array
operations can be performed. During port login, two N_Ports (devices) swap
service parameters and make themselves knoWn to each other.
" Process login is also known as PRLI. Process login is used to set up the
environment between related processes on an originating N_Port and a responding
N_Port. Fig. 2.12.1 Zoning
Fabric login : Performed between an N_Portand an F_Port. To log on to the fabric, Zoning provides access control in the SAN topology. Zoning defines which HBAs
a node sends a FLOGI frame with the WWNN and WWPN parameters to the can connect to which targets. When you configure a SAN by using zoning the
login service at the predefined FC address FFFFFE (Fabric Login Server). In turm, devices outside a zone are not visible to the devices inside the zone.
the switch accepts the login and returns an Accept (ACC) frame with the assigned
FC address for the node. Immediately after the FLOGI, the N_Port Zoning has the following effects :
registers itself 1. Reduces the number of targets and LUNs presented to a host.
with the local Name Server on the switch, indicatng its WWNN,
WWPN, port 2. Controls and isolates paths in a fabric.
type, class of service, assigned FC address and so on. After the N Port has logged
in, it can query the name server database for
information about all other logged in 3. It can be used to separate different environments, for example, a test from a
ports. production environment.
Reasons for zoning
2.12 Zoning 1. Zoning reduces or eliminates cross talk between initiator HBAs.
FC or FCoE port
" SAN zoning is a method of arranging 2. Zoning reduces the number of available paths to a particular
over the physical configuration of the Fibre Channel devices into logical groups and reduces the number of paths between a host and a particular LUN that is
fabric. SAN zoning may be utilized to
visible.

knowledge TECHNICAL PUBLICATIONS- An up thrust for knowledge
Storage Networking Technologe
snd Management Storage Networking Technologies
InformstionStorage limited access between
2.12.1 Types of Zoning
3. Zoning
increases
nodes ofa SAN.
security because there is
aif eren Zoning are of three types :
isolating problems that occur and
4. Zoning resolution time by
by
improves SAN reliability limiting the problem space. helps t 1. Port zoning 2. WWN zoning 3. Mixed zoring
reduce problem zoning is usually performed within the
In switched fabric fibre channel topologies, smaller subsets for the Types of zoning
"
Switch. It involves
partitioning a fabric into example, when purposes o
restricting interference and
increasing security, for establishing a
P irewalL.
zone can be active at a any time. S:
multiple zones but only one
" Fabric supports member may be th Port zoning wWN zoning
is a set of member. One Mixed zoning
of zone is called zone set. Zone a hierarchy of members. zon
process defined
of more than one zone. Zoning Fig. 2.12.3 Types of zone
2.12.2.
zone sets. It is shown in Fig. Port Zoning :
Zone set
" Port zoning is also called hard zoning. Two or more nodes must be a part of same
zone receive route information and to communicate.
Hard zoning can restrict the traffic. It is enforcement based zoning. Port zoning
does not allow zones to overlap.
Zone 1 Zone 2 Zone 3 Zone 4 . The advantage of hardware zoning is that it can be implemented into a routing
engine by filtering
WWN Zoning :
WWN zoning is also called as soft zoning. It is based on the information. Zones
Zone Zone Zone Zone Zone Zone
Zone
member member member mernber mernber member member
are defined by using World Wide Names.
Two or more nodes must be parts of same zone receive route information to each
other. It will not restrict the trafic.
Fig. 2.12.2 Zone with zone members
" A major advantage of WWN zoning is the ability to re-cabled the fabric without
" Group of zones is called zone set. Zone sets are also referred to as zone having to redo the zone information.
configurations. " WWN zoning is susceptible to unauthorized access, as the zone can be bypassed if
" Characteristics of Zones :
an attacker is able to spoof the World Wide Name of an authorized HBA.
1. Zone can vary in size.
Mixed Zoning :
2. Devices can belong to more than one
zone. " It is combination of port zoning and WWN zoning. It is also called hybrid zoning.
3. Members in a zone can access each
other
" Zone sets provides
following advantages : 2.13 Fibre Channel Topologies
1. Only one zone set can be
active at any time. T'opology is usually described in terms of how the switches are interconnected,
2. All zones in a zone set can
all switches in the fabric.
be activated or
deactivated as a single entity acros such as ring, core-edge, and edge-core-edge or fully meshed.
3. A zone can be a
member of nore than one zone set.

knowledge
Management
Informstion Storage and Storage Networking Technologies
2.13.1 Mesh Topology
topology". This topology support FC switch 1
may be interconnected in a"meshed
" Switches devices in the mesh. FC Switch 2
muliple paths to all the number of nodes that Can Server Director
network increases, the
number of ports on a mesh topoloy
" When the increases. Fig. 2.13.1 shows FCswitch 3
and communicate also
participate partial esh. Storage array
the two types: Full mesh or Edge tier Core tier
A mesh topology are of
Fig. 2.13.2 Single core
FC switch 1 FC switch 2 topology
In single core, al hosts are
Server to the core tier.
connected to the edge tier and all storage is connected
FC switch 3FC switch 4 Storage array
Fig. 2.13.3 shows dual core topology.
Fig. 2.13.1 (a) Partial mesh
FC Switch 1
Director 1
FC switch 1 FC switch 2 FC switch 2

Server
Server FC switch 3 Director 2
FC switch3 FC switch 4 Storage array

Storage array Edge tier Core tier
Fig. 2.13.1 (b) Full mesh Fig. 2.13.3 Single core topology
. Full mesh topology : In this topology, every switch is connected to every other 2.13.3 Benefit and Limitation of Core-edge Fabric Topology
switch. This type of topology is suitable when the number of switches involved js Benefits:
small. Even number of switches is used while designing fullmesh topology. 1. It provides one hop storage access to all storage.
" In a partial mesh topology, several hops or 1SLs may be required for the traffic to 2. Used in larger environment.
reach its destination. Hosts and storage can be located anywhere in the fabric, and
storage can be localized to a director or a switch in both mesh topologies. Odd 3. The servers and storage are on different switches, this topology provides ease of
numbers of switches are used while designing partial mesh topology. managemnent as well as good performance.
Limitations : Storage and core connections are in contention for expansion. In
2.13.2 Core-Edge Fabric other words, this topology allows for only minimal growth.
" Core-Edge is a reference topology widely used in the layout of SAN fabric. 2.14 Network Atached Storage
" Hosts are attached to smaller port count 'edge' switches, which are attached to the
" NAS stands for Network Attached Storage. NAS is a file level computer data
core switch via ISLs. This topology can grow to hundreds of ports while ensuring
that hosts only have to traverse two switches to access their storage. storage server connected to a computer network providing data access to a
by its
Hosts that
require lower latency or higher throughput can be connected directly to the core heterogeneous group of clients. It is specialized for serving files either
switch. hardware, software, or configuration.
file-based storage
" The edge tier NAS devices are storage arrays or gateways that support
contains switches and offers an inexpensive approach to CiFS, and are typically connected via an IP network.
more hosts in a fabric. Fig. 2.13.2 shows adau's Protocols such as NFS and
single core topology.
TECHNICAL PUBLICATIONS- An up thrust for knowledge TECHNICAL PUBLICATIONS- An up thrust for knowledge
Storage Nel
2-34 lbgies laformation Storage and Management
Management 2-35 Storage Networking Technologies
Storage and connectivity,
Information
methodologies for storage configurability. either 2.15 Benefits of NAS
support two further
Most NAS
devices will provide even
fabric attach, which
direct attach or
1. NAS Support NFS and CIFS file system.
utilization options. and hierarchical interface. NAS 2. Entry level NAS systens are quite inexpensive.
typed, variable-size
provide a richer, magnetic media, usually through
a 2 NAS ensure security and user authentication.
NAS systems interface with non-volatile
systems internally ensure data reliability. A Tt helps for easy and fast communication.
SAN-like interface to
contain u
file system. NAS Systems
network between client and
as much metadata a 5. NAS appliances already include integrated mechanisrns for backup, data
" NAS places a interprets metadata, so they do not send the synchronization and data replication.
file sys-tem that trusted less, mainly operating
network. NAS systems have traditionally these clients. A Simple to administrate, even remotely, i.e. from another location.
the but not the users
of
systems of their clients within the server but can exist
7 NAS device may support one or more RAID levels to
make sure that individual
be located
device does not need to networked NAS devi disk failures do not result in loss of data.
"A NAS
LAN and can be made up of multiple
anywhere in a hard disks, often arranged into logical, 8. NAS is a much simpler and less expensive technology.
contain one or more
NAS systems usually RAID arrays.
redundant storage containers or server-centric SvSto 2.16 Network File Sharing
any of the activities that a server in a
" NAS does not provide NAS allowe
such as e-mail, authentication or file management. uilizos
.A file can be informally defined as a collection of data, which can be logically
typically provides, network that already viewed as a stream of bytes.
to be added to a
more hard disk storage space upgrades.
down for maintenance and Afile is the smallest unit of storage in the UNDXfle system.
servers without shutting them arranged into logical
NAS systems contain one or more hard disks, often The file system has a hierarchical structure with its highest level directory called
" RAID arrays root (denoted by ). Immediately below the root level directory are several
redundant storage containers or
subdirectories, most of which contain system files.
NAS Devices
2.14.1 General Purpose Servers vs 2.16.1 Accessing a File System
function. It supports operation like
" NAS device is optimized for file serving
storing, retrieving and accessing file for client
and application. File system is mounted before use of it. During boot process, OS mounts a local
purpose server runs on any file system.
" NAS device is dedicated to file serving. General to
operating system. The mount system call connects the file system in a specified section of a disk
system hierarchy, and the unmount system call disconnects a file
the existing file
" NAS device has a real time OS and it uses open standard protocol.
single hardware system from the hierarchy.
" Like traditional file servers, NAS follows a client/server design. A access data in a disk section as a file
head, acts as the interface between the The mount system call allows users to
device, often called the NAS box or NAS
NAS and network clients. system instead of a sequence of disk blocks.
" File systems created on other
devices can be atached to the original directory
These NAS devices require no monitor, keyboard or mouse. They generally run an mount is established, users are
hierarchy using the mount mechanism. When
embedded operating system rather than a full-featured NOS. One or more disk
unaware of crossing mount points.
drives can be attached to many NAS systems to increase total capacity. Clients structure called the mount table that allows
always connect to the NAS head, however, rather than to the individual storage Kernel maintains a system-wide data
single directory hierarchy.
devices. multiple file systems to be accessed via a in a collection.
lists files and subdirectories
Drectory is a data structure that
entries one for each ile.
Directory structure contains a list of
up thrust for knowledge
TECHNICAL PUBLICATIONS - An
Information Storage and
Menagement
2-36
yTecnnologiss Information Storage and Management
2-37
The root file system has
directory /. " Network File Systen (NFS) is
The root is identified by the
subdirectories. UNDX and
Linux based file system uses slash (/) for
specifyingseveralroot and Common Internet File
protocol.
standard for
Systern (CIFS) isUNIX client-server file sharing protocol
standard
directory. Iroot Windows
operating system
contain The NFS client maintains a cache of file and
"A directory can
settings will not ensure that files created or directory
modified on one attributes.
A The
any number of files.
directory can also contain on another system within a minute of file
creation or systern will be default
visible
subdirectories. Because a Network file system is common distributed
fle
modification.
directory is nothing more bin etc dev Usr
vmunix . Email and FTP is used for sending sharing system.
than a special kind of file, information
protocols will not support true file sharing. to remote users. But
these two
directories also have
Passwd local
Domain
to map the host name to network address translation forName Systen (DNS) is used
names. Passwd
the Email and FIP. The DNS is a hierarchical the internet. DNS
replace
" Fig. 2.16.1 shows UNDX Fig. 2.16.1 UNIX directory structure Industry is moving towards the new technology. Sonaming
distributed
system.
directory structure. new protocol called
leaf nodes Iightweight Directory Access Protocol (LDAP) is used for secure
The tree consists of files and directories, with the files at the naming service. distributed
about the files and other directories L.
directory contains name information
IDAP allows user to search for an individual
resides in it.
located.
without knowing where they are
2.16.2 Network File Sharing An
LDAP directory can be distributed among many
servers. Each server can have
a replicated version of the total directory that is
. Files are shared between numbers of users. File protection, naming and sharing is synchronized periodically.
e An LDAP server is called a Directory System Agent. An
issue for ile sharing. Directory structure may be allows files to be shared by LDAP server that receives
users. Sharing must be done through a protection scheme. a request from a user takes responsibility for the request, passing it to
other DSAs
as necessary, but ensuring a single coordinated response for the user.
" Afile is logically a container for data. Users may create a file and store data by
writing it to the file. UNDX file does not contain the end of mark. File size is not 2.17 Components of NAS
stored in the file and even its name.
" In UNDX operating system, file owner can perform all operation on a file. Group " NAS Consists of following components :
1, CPU and memnory
member can execute one subset of those operations and other than the group user
can execute another subset of operations. 2. NIC
" When a user requests an operation on a file, the user ID can be

compared with 3. Operating system
the owner attribute to determine if the regquesting user is the 4. NFS and CIFS
owner of the file.
Likewise, the group IDs can be compared. 5. Storage protocol.
Computer network is used to support remote file system. User can access " Fig. 2.17.1 shows NAS components.
file system. File remote
Transfer Protocol (FTP) and Secure Shell protocol (SSH). Other " NAS uses industry standard storage protocols (ATA,SCSI and FC) for connecting
supporting protocol like distributed file system and world wide
for remote file system. web is also used and managing physical disk resources. Network file system and Common Internet
File System protocols used for file sharing. Network interface card is used for
Client-server model allows clients to mount remote file providing network connectivity.
server is connected with multiple clients and it systems from servers. The
serve to all clients. Client arnd
user-on-client iderntification is insecure or complicated. " NAS devices are the most appropriate for high-demandengineering applications
like computer-aided design and software development.
TECHNICAL PUBLICATIONS - An up thrust for TECHNICAL PUBLICATIONS - An up thrust for knowiecye
knowledge
Storage NetworkKing Technologies
Management
2-38 Information Storage and 2-39
Management Storage Networking Technologies
Informetion Storageand
is no awareness of disk
There volumes or disk
sectors. File
protocol, such as : NFS andsystern
Server 2
Server1 remotely using a network file access is mounted
shows NAS I/Ooperation. CIFS. Fig. 2.18.1
Client machine NAS device

Storage array
IP network
Storage
interface Storage
i interface
iNetwork Network
interface i
Tinterfacei
Interface for network
NFS CIFS Fig. 2.18.1 NAS VO operation

Operating system
Interface for storage follows :
" Steps are as
1 NAS device receive the I/0 request from the network. User sends the number of
I/O request by using TCP/P protocol.
2. NAS device convert the user I/O request into required physical storage request.
This request is block level I/0 request. After converting request, operation is
Storage arrays performed on the physical storage pool.
This device
3. Physical storage pool sends processed data back to the NAS device.
converts into appropriate format.
the client.
4. Using TCP/IP protocol, NAS device delivers this packet to
Fig. 2.17.1 NAS components 2.19 NAS Implementation
network and you want both
" If you have both UNIX and windows users on your
NAS devices can
NAS is implemented in two ways : integrated and gateway.
groups to be able to share files, NAS devices are most suitable.
make use of existing directories of user accounts fromn Windows, Netware or 2.19.1 Integrated NAS
UNX Server. NAS so it makes the integrated
Integrated NAS contains all the components ofshows integrated NAS. Integrated
Fig. 2.19.1
2.18 NAS /O Operation NAS a self contained environment. integrated NAS which has its own
looking at an
NAS is useful when customer is
" In NAS, all I/0 operations are performed at file level I/O protocols. The NAS storage. device to high end solution.
workload consists of only I/0 requests and no metadata requests, i.e., no creaie integrated NAS solution is available with low-end systems
delete, change attribute operations. Using I/0 to perform metadata operations call An and communicate with file
are connected to the local area network (TCP/IP).
increase performance in certain instances. 1ney potocol/nternet protocol
through trans-mission control
knowledge
PUBLICATIONS- An up thrust for
TECHNICAL
2-40
Storage Networking Technologle,
Management Information Storage and Management
InformationStorage end 2-41
Windows operating
Storage Networking Technologi s
systems client Servers
Integrated network
attached storage
Unix system client
NAS
Fig. 2.19.1 Integrated
Ethernet Ethernet
2.19.2 Gateway NAS switch LAN
internal processor but without integrated

" It is NAS device with the
Instead, the NAS device connects to storage by direct
attachment or by a SAN storage.
" In a NAS gateway, the proprietary RAID controller is replaced with an open, NAS NAS
ANSI-standard SAN interface, usually a fibre channel host bus adapter. This gateway gateway
allows the NAS gateway to use shared, nonproprietary storage, which provides FC
the flexibility for an effective and scalable storage strategy. Fiber
. When deploying NAS gatewas, it is important to select the right shared storage channel
Switch
platform to house your organization's data.
FO
" Fig. 2.192 shows NSA gateway.
" Afront end to disk storage that is highly scalable. A NAS gateway functions like
a diskless file server with urnlimited storage. Unlike the traditional NAS device.
which contains the file sharing protocol for access, the file system for storage and
the disks themselves, a NAS gateway contains only the file sharing protocol and
the file system.
RAID array RAID array
Irtegrated network
FC Fig. 2.19.2 (b)
Switch
atached siorage
Fibre " The front end of the NAS gateway connects to the LAN over Ethernet. The back
channel end connects to a RAID array via Fibre Channel or to a Fibre Channel switch, in
SAN which case the NAS gateway becomes a front end to a storage area network.
UNDX system client
Windows operating Gateway NAS Connectivity
systerns client Communications between the NAS gateway and the storage system in a gateway
solution is achieved through a traditional FC SAN.
Fig. 2.19.2 (a) Gateway NAS
1o deploy a gateway NAS solution, factors, such as multiple paths for data,
redundant fabrics, and load distribution must be considered.
TECHNICAL PUBLICATIONS- An up thrust for knowledge TECHNICAL PUBLICATIONS - An up thrust for knowiedge
Management
2-42
NAS
nologies Information Storage and Management
2-43
Infometion Storage and
Integrated NAS and Gateway 2.20.1 NFS Storage Networking Technologies
2.19.3 Difference between Gateway NAS
" Network File System (NSF) is
system independentdeveloped by Sun
Integrated NAS Gateway NAS uses both Fibre Channel and distributed file
Integrated NAS uses

Fibre Channel
disks only. SATA disks.
e NFS uses remote procedure
of the
call
hardware andMioperating
crosysterns to provide
systern.
a
Gateway NAS relies on storage array mechanism is independent of the nechanisrn for rernote file operations. RPC
Virtual File System (VFS) : File transport protocol.
devices management software to manage storage
storage
Integrated NAS manages
Aifferent file systerns. Requests for system
devices.
through NAS software.
systems.
interface that allows NSF to support
It uses generic operating NFS. operation on remote files are routed by VES to
operating systems.
It uses specialized Gateway NAS allows only block level access to Requests are sent to the VFS on the
file and block level remote using the Rernote Procedure Caln
Integrated NAS allows both
the clients.
(RPC), and External Data
initiates files system operationRepresentation
access to the clients. (XDR). VES on the rernote server
Scale Out NAS
Connectivity locally. VES interface is based on a structure called
a virtual node (vnode).
2.19.4 Scale Out NAS and
as a single NAS J. NES uses client caching to reduce
together in a cluster that works
" It pools multiple nodes network load. NFS is uilt on the e-mail
Pooling is managed by centrally. addition of nodes to Hhe pod top of a number of protocols, as NFS www rlogin
Scales performance and/or capacity with shown in Fig. 2.20.1 in the
" single file system that runs on all nodes :
non-disruptively. It creates a following alphabet soup diagram. RPC
TCP
cluster. . In the NFS, file blocks are cached
it can access entire file system. The A.
" Clients are connected to any node and on demand.. Since both clients and UDP
are added.
system grows dynamically as nodes servers cache file blocks and
parity protection.
" Stripes data across all nodes in a pool along with mirror or metadata, the main advantage of
Scale-out NAS Connectivity
this approach is that many file Packet radio
for back-end and operations (e.g., read, and write) Ethernet
" The clusters use separate internal and external networks
ATM
can be done locally. However, NFS
front-end connectivity, respectively. The internal network provides connections fo
Fig. 2.20.1
has to worry about handling
intra-cluster commurication.
failures and maintaining cache consistency.
" The external network connection enables clients to access and share file data
NFS employs read ahead and large block size for data transfers to improve the
Each node in the cluster connects to the internal network.
sequential read performance.
" The internal network offers high throughput and low latency and uses high-spe:
networking technology such as Infiniband or gigabit Ethernet.
A Network File System (NFS) allows remote hosts to mount file systems over a
network and interact with those file systems as though they are mounted locally.
To enable a client to access a node, the node must be connected to the extern All versions of NFS can use Transmission Control Protocol (TCP) running over an
Ethernet network. Redundant internal or external networks may be used for hig IP network, with NFSV4 requiring it.
availability. " NFSv2 and NFSV3 can use the User Datagram Protocol (UDP) running over an IP
2.20 NAS File Sharing Protocol network to provide a stateless network connection between the client and server.
" NAS devices support multiple file service protocols to handle file I/O requeskst Network File System (NES) is standard for UNIX client-server ile sharing protocol
a remote file and Common Internet File System (CIFS) is standard Windows operating system
system. NFS and CIFS are the
common protocols for file
sharing Protrocol. The NFS cliernt maintains a cache of file and directory attributes. The
knowledge
2.20 Connectivity
Scale-out
NAS 2.19.4 access 2.19.3Information
" " " through
Integrated
NAS Integrated
" " " " " Integrated uses Itsoftware.
Ethernet
NASa availability. networking intra-cluster
Each
To Thecommurication. The front-end The Stripessystem non-disruptively.
Clientscduster. ScalesPooling It to
remote NAS enable pools Scale specialized
the Storage
devices internal node
external NS NAS Difference
clusters clients. NAS
data grOws are performance multiple is uses NAS and
ile File network. a connectivity,
in managed Out allows manages Integrated
technology across connected operating
system.
support client network the use dynamically Fibre Management
Sharing network NAS both
cduster t nodes storage
between
ECHNICAL Redundant to separate all Channel
NFS creates by and e
flsystems.
multiple
access offers respectively. nodes to and/or centrally.
connection
such connects together and devices
TIONS- and Protocol asany Scale disks Integrated
a as high internal in nodes node a bl ock
CIFS file node,
internalInfiniband a capacity
single only.
to pool in Out level
service throughput The 2-
42
are enables
the are and a
the and file cluster NAS devices. NAS
An the or internal
along added. clients.
Gatewayth e managenment
It Gateway SATA
disks.
Gateway
up internal it with uses
protocols
thrustcommon
external node or external can system and
clients with Connectivity
gigabit and that generic
for must network. access addition NAS NAS NAS
wledge low network mirror that works Gateway
networks to softwarerelies
protocols to Ethernet.
be networks allows operating usesGateway
NAS
handle latency access entire runs TechnologlexStorage
connected or of on both
provides as only to
may parity a storage NAS
and and file nodes
on systems. manage Fibre Networking
for file for single block
be protection. system. all
file I/O to uses share
connections array Channel
back-end nodes to NAS level storage
sharing used the
requests high-sped file the
external
for The accoes.
in device ,
hign data.
for and file the nonl
" " " 2.20.NFS1Information

"
network A
sequential
Pperformance. operations
read NFSfailures this serverS
has can metadata, on
following network
In shown intop a(RPC),
NFS initiates NFS. Virtual
OtoCo. Network le NSV2
and Fietwork virtual different
Requests systems. mechanismdistributed NES Network
network, versionsNetwork l be demnand. the Storage
employs toapproach
ofa uses uses
Common and NFS,
to and and done NFS node files and file File and
The handling
worry
about the
cache alphabet number load.
client are remote
is file File
provideNFSV3 with of File maintaining locally. (e.g., Since file 2.20.1 Fig. External
system Systerm Management
NES System interact read (vnode). sent to independent
systemSystem
Internet NFS main
is blocks caching
NFSv4 System read, file
that both soup ofuilton procedure
L client acan can ahead However, is operation
stateless
(NES) with advantageblocks protocols, Data the Requests (VFS) independent (NSF)
File use cache and many are
clients diagram.
requiring use (NFS) to VFS
aintains the those and cached thein reduce Representation : of
S- System (CIFS) is Transmission write) for File the call is
standard network User allows consistency.
large NFS locally. on
file as the transport developed
43
2-
file ofand and theoperationsystem
it. nechanisn of
a Datagram the
systemsremote block
NFS VES renote
Ancache connection for
up Control Ethernet interface(XDR). interface protocol. hardware by
hrust UNIX
is size UDP RPC
on
of as
hosts using remote for Sun
for standard Protocol though for
fle
iecge cient-serverbetween Protocol Www isVFS that rermote and Technologies
to data the Microsysterns Networking Storage
and 2.20.1
Fig. based files operating
(UDP) they mountile ATM on
Remote allows
Windows
directory transfers file
the (TCP) e-mail on are
are TCP the
runing
file client a routed NSFoperations.
systen.
running Packet
radio renmote
Procedure
structure to
attributes.
operatingsharing and mountedsystems
over a to
improve by to provide
ove riogin support
protocol server. VFS
a
overlocaly. calledserver RPC
system Call to
Ine an the a
Storage
Networking
Technologies less with attached because varying
it supports improvng awitn
cornect system by block
disk
are equipped block
which adapters file
logical
in
data device it
todaycurrently havestart
organization
to
a thus
the require by
SAN topologies
are the
clients manage storage SANs
SAN thatChannel
block
copy
datatransfers Channel
operation.
mirrorS fast, absence.
peer
devices
servers
the implementingthe know/edge
Fibre addresses
and
Fibre digital s works.on The
iFC
nultiple class
and
uses
servers allowing
topologies. (SAN)
andto
aSAN. SAN Backups numbers,
Server SAN blocks.SAN Answers dedicated applications.
the Networkforthrust
SCSI The
for in all thus needed.
capabilities
tradeoffs,
capacity.between SAN 10Gbps
by up
2-45 a
utilize
to head technology accessed
different PUBLICA
An
and name a Area
deployment
and
maintenance.
connect
NAS
on or
generated data with ? is ? today's as
sharing sharing NAS may (DAS) SAN & up
file storage 4 Storage
LANto the Questions cable
leading 2, be several scale
byfile Networks for of with to TECHNIC
fles times. between withprotocol by are datatransfers Storage requirements DAS. include
file NAS managed attached a Channel
via devices and
Storage
Information
Management
mirrors the 1Gbps supports to
of low-volume
response connected PC the setup due
addition HITP identifies TCP/IP of advantages
Comparison blocks offset,
files,
not Attached is storage cost
to "High
advantagesavailability.
is and Marks
direct or Fibre Channelto rate
andi r
1CIFS system server It reliable
Simplified to Machine Backups NASbytemetadatauses the up access
use reliability.
for
:Advantages sensitive NFS,NAS File NAS TwO is Direct meeting
of allows Ievels
of
Easy Best unit
A
and What ato Why Fibre speed ListDAS High
small,
2.22.1 No.Sr. :
2.23 0.1 Ans.
directly
:Ans. It :
Ans.
1. 2. 3. Q,2 " Q,3 "
Storage
Networking will
modification
Technologleg file the NAS latency or
svsterm modified, directory,
system CIES request file-sha affect
of
protocol. appliance
native connections. sources
one to IP
or protocol with Active
overwriting. are
on creation a significant
modifiedsystem. sharing was and integrity contents associated
: LDAP,
this andembedded restore
file sharing knowledge
file use from directory most
or of OSs data issues as
network. automatically
created
minute a systems server users the such requestNAS distances.
for
file is in ensure Performance
latency are
thrust
distributed (CIFS) used or of
a and
prevent service
performance up
2-44 files a Clientover file one of
metadata An
within System
also to and Advantages -
that System workstation features
to
can When is directory switches access. long PUBLICATION
protocol.
systemns is congestion
bandwidth
ensure common CIFS locking and NAS and over
system : NAS base CIFS.
File File following tolerance notification andlookup transfer TECHNICAL
server 2000. a
not
Management application in record Affecting and
be is lnternet
visible
another Internet
usedWindows
protocol. Network
network, affect with
environment. router and data
will system hopes Directory NES data
from
theand fault server
isnotified. only
typically that AuthenticationRetransmissionApplication
3.
settings Commonservices provides change of utilized in of
and on file Common client/server file TCPsupports factors readsharing
in IP
Factors performance. Numbers Small-block
Storage Network uses uses uses NAS or Applications
: Limited
default access
is protocol File Other NIS Over File
The CIFS CIFS It It It NAS a File
Information 1. 5.
2.20.2 a. b. c. d. 2.21 in 2. 4. 2.22
" " " " " 1. 2 3.
Information Storage and Management 2-46 Storage Networking Technologa
Elimination of network setup complications.
Storage capacity expansion.
" Data security and fault tolerance.
Q.4 Define SATA.
Ans. : Serial ATA is a standard for connecting and transferring data from
drives to computer systems. It provides point-to-point connectivity. hard disk
Q5 What is Fibre channel ?
Ans. : Fibre channel is a high speed network technology that runs on
optical fiber cables and serial copper cables. high speed
Q.6 What is SAN management software ?
Ans. : SAN management software organizes the connections, storage
elements and
computer systems in a SAN storage system. SAN management software does this so
that data transfer is secure and robust. SAN
management software can also help with
plarning and monitoring of storage systems. The software can be provided for large
enterprises, mid-sized businesses and small comparies. The software may be ofered as
a commercial product or as an open-source or free
solution.
Q.7 Define Zoning.
Ans. : SAN zoning is a method of arranging Fibre
Channel
over the physical configuration of the fabric. SAN zoning devices into logical groups
may be utilized to
compartmentalization of data for security purposes. Each device in a SANimplement
may be
placed into multiple zones.
0.8 List characteristics of Zones.
Ans. : Characteristics of zones are :
1. Zone can vary in size.
2. Devices can belong to more than one Zone.
3. Members in a zone can access each other
Q.9 List the components of NAS.
Ans. : NAS components are CPU and memory ,
CIFS, Storage protocol. NIC, perating system , NFS and
thrust for knowledge

ST Unit 2

Uploaded by

Copyright:

Available Formats

ST Unit 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ST Unit 2

Uploaded by

Copyright:

Available Formats

Infomation Storage and Management 2-1 Storage Networking Technologies

UNIT-II Mind sharing

Syllabus DAS Benefits and Limitot

TECHNICAL PUBLICATIONS- An up thrust for knowledge

2.1 Direct-Attached Storage

SATA Type Bandwidth Bus Speed Signal Rate

built into the drive. SCSI-1 5 MB/S 8

TECHNICAL PUBLICATIONS - An up thrust for TECHNICAL PUBLICATIONS - An up thrust for knowledge

1. Network resource used exclusively for the storage.

hub device for

TECHNICAL PUBLICATIONS An up thrust for knoedodee

Arbitrated Loop . In a switched fibre channel fabric,

FC -2 Framing, flow control, zoning, ordered sets

4. F Port : A fabric port is used to connect an N_Port to a switch that is not

TECHNICAL PUBLICATIONS- An up thrust for TECHNICAL PUBLICATIONS - An up thrust for knowledge

Information Storage and

. WWNs are similar to the Media Access Control (MAC)

c. Class-3 service : It is also a connectionless service based on frame switching Supported

TECHNICAL PUBLICATIONS- An up thrust for

TECHNICAL PUBLICATIONS- An up thrust for knowledge

FC switch 1 FC switch 2 FC switch 2

Server FC switch 3 Director 2

FC switch3 FC switch 4 Storage array

" When a user requests an operation on a file, the user ID can be

Client machine NAS device

Interface for network

NFS CIFS Fig. 2.18.1 NAS VO operation

Unix system client

internal processor but without integrated

Integrated NAS uses

" " " 2.20.NFS1Information

You might also like