What Is The Hard Disk Architecture
What Is The Hard Disk Architecture
What Is The Hard Disk Architecture
Storage Area Network, SAN is a block level data storage which enables the communications between
initiators and targets over fiber channel network using FC protocols.
What is NAS
Network Attached Storage, NAS is a file level data storage which provides shared data access to multiple
group of servers using TCP|IP protocols.
What is DAS
DAS is a Direct Attached Storage, where the servers will directly get connected to the storage array. This is
a traditional way of implementing SAN.
The architecture of a hard disk consists of several physical components that include:
Platters
Spindle
Read|write heads
Tracks
Sectors
Platters
Hard disks are organized as a concentric stack of disks. An individual disk is referred to
as a platter.
Spindle
The platters within the hard disk are connected by a spindle that runs through the
middle of the platters.
The spindle moves in a unidirectional manner along its axis (either clockwise or
counterclockwise).
Read|write head
Each surface on a platter contains a read|write head that is used to read or write data
onto the disk.
The read|write heads can move back and forth along the surface of a platter. Read|write
heads are in turn connected to a single actuator arm.
Tracks
Each surface of a platter consists of a fixed number of tracks. These are circular areas on
the surface of a platter that decrease in circumference as we move towards the center of
the platter.
Sectors
Each track is divided into a fixed number of sectors. Sectors divide track sections and
store data.
DISK TECHNOLOGIES
IDE, SATA, SCSI, SAS, FC, SSD HARD DISK TYPES
IDE | ATA
================
IDE (Integrated Drive Electronics), originally meant to refer to the hard disk drive that integrates the
controller and the disk body, is a hard disk transmission interface. There is another name called ATA
SATA disk
==================
SATA (Serial Advanced Technology Attachment) port hard disk is also called serial hard disk. SATA is
named after its serial data transmission method. In the process of data transmission, the data line and
the signal line are used independently, and the transmission clock frequency remains independent, so
compared with the previous PATA, the transmission rate of SATA can reach 30 times that of parallel . It
can be said that SATA technology is not an improvement of PATA technology in a simple sense, but a
SCSI disk
==================
The English full name of SCSI: Small Computer System Interface. The reason for its appearance is
mainly because the hard disk speed of the original IDE interface is too slow and the transmission rate is
too low, so high-speed SCSI hard disks appear. In fact, SCSI is not specifically designed for hard disks,
SAS disk
==================
SAS (Serial Attached SCSI) is serial attached SCSI, which is a new generation of SCSI technology. Like
the popular Serial ATA (SATA) hard drives, they all use serial technology to achieve higher transmission
speeds, and improve internal space by shortening the cable. SAS is a new interface developed after the
parallel SCSI interface. This interface is designed to improve the performance, availability, and
expandability of the storage system, and to provide compatibility with SATA hard drives.
SAS interface technology can be backward compatible with SATA. Specifically, the compatibility of the
two is mainly reflected in the compatibility of the physical layer and the protocol layer.
FC
==================
The English spelling of FC is Fiber Channel. Like the SCIS interface, Fiber Channel was not originally
an interface technology designed and developed for hard disks. It was specifically designed for network
systems. However, as storage systems demand speed, they are gradually applied to hard disk systems.
Fiber Channel hard disks were developed to improve the speed and flexibility of multi-disk storage
systems. Its appearance greatly improves the communication speed of multi-disk systems. It uses fiber
SSD|EFD
==================
Solid State Disk (Solid State Disk or Solid State Drive), also known as electronic hard disk or solid state
electronic disk, is a hard disk composed of a control unit and a solid state storage unit (DRAM or FLASH
chip).
RAID Concept
RAID Levels
Whether hardware or software, RAID is available in different schemes, or RAID levels.
The most commonly levels are RAID 0, 1, 5, 6, and 10. RAID 0,1and 5 work on both
HDD and SSD media. (RAID levels 4 and 6 also work on both media, but are rarely seen
in practice.)
Raid 0: Striping
Requiring a minimum of two disks, RAID 0 splits files and stripes the data across two
disks or more, treating the striped disks as a single partition. Because multiple hard
drives are reading and writing parts of the same file at the same time, throughput is
generally faster.
RAID 0 does not provide redundancy or fault tolerance. Since it treats multiple disks as a
single partition, if even one drive fails, the striped file is unreadable. This is not an
insurmountable problem in video streaming or computer gaming environments where
performance matters the most, and the source file will still exist even if the stream fails. It
is a problem in high availability environments.
RAID 1: Mirroring
RAID 1 requires a minimum of two disks to work, and provides data redundancy and
failover. It reads and writes the exact same data to each disk. Should a mirrored disk fail,
the file exists in its entirety on the functioning disk. Once IT replaces the failed desk, the
RAID system will automatically mirror back to the replacement drive. RAID 1 also
increases read performance.
It does take up more usable capacity on drives, but is an economical failover process on
application servers.
RAID 5 stores parity blocks on striped disks. Each stripe has its own dedicated parity
block. RAID 5 can withstand the loss of one disk in the array.
RAID 5 combines the performance of RAID 0 with the redundancy of RAID 1, but takes
up a lot of storage space to do it – about one third of usable capacity.
This level increases write performance since all drives in the array simultaneously serve
write requests. However, overall disk performance can suffer from write amplification,
since even minor changes to the stripes require multiple steps and recalculations.
RAID 6 offers higher redundancy than 5 and increased read performance. It can suffer
from the same server performance overhead with intensive write operations. This
performance hit depends on the RAID system architecture: hardware or software, if it’s
located in firmware, and if the system includes processing software for high-performance
parity calculations.
RAID 10: Striping and Mirroring
RAID 10 requires a minimum of four disks in the array. It stripes across disks for higher
performance, and mirrors for redundancy. In a four-drive array, the system stripes data
to two of the disks. The remaining two disks mirror the striped disks, each one storing
half of the data.
This RAID level serves environments that require both high data security and high
performance, such as high transactional databases that store sensitive information. It is
the most expensive of the RAID levels with lower usable capacity and high system costs.
If one or more
Data is split evenly
Large size and the drives fails, this
RAID 0 Striped disks between two or more No redundancy.
fastest speed. results in array
disks.
failure.
Two or more drives A single drive failure Speed and size is Only one drive is
RAID 1 Mirrored disks have identical data on will not result in data limited by the slowest needed for
them. loss. and smallest disk. recovery.
FC Layers
What is Fiber Channel
FC (Fiber Channel) is a network technology, predominantly used within storage area networks, to provide
high-speed, loss-less delivery of raw block data between computer data storage and server devices.
The Fiber Channel network (aka fabric) is a dedicated high-speed, low latency storage network, supporting
bandwidth speeds of 2, 4, 6, 8 and 16Gbps, connecting:
HBAs (Host Bus Adapter) - Dedicated server storage adapters (aka the Initiator)
Storage Systems - Contains storage controllers, and disks (aka the target).
FC Switches - High-speed Fiber Channel switches.
The FCP (Fiber Channel Protocol) operates over the FC fabric. Predominantly FCP is the implementation
of SCSI over an FC network, or in other words, SCSI data is encapsulated and transported within FC frames.
FC Layers
When it comes to learning the various layers of Fiber Channel it is important to understand that FC does not
follow the traditional OSI 7 layer model. Instead, it is broken down into 5 layers. As shown below:
FC-0 - Defines the physical media used to link two Fiber Channel ports, including cabling types, optical and
electrical parameters for a variety of data rates, maximum transfer distances, and noise limits. Fiber Channel
supports two types of cables: Copper and optical.[2]
FC-1 - Defines the transmission protocol including serial encoding and decoding rules, special characters and
error control.[3]
FC-2 - Defines the transport mechanism of Fiber Channel and the framing rules of the data to be transferred
between ports, the different mechanisms for controlling the service of classes and the means of managing the
sequence of a data transfer.[4]
FC-3 - Defines common services required for advanced features such as striping, hunt group, and multicast.
FC-4 - Defines the application interfaces that can execute over Fiber Channel. It specifies the mapping rules
of upper layer protocols using the FC levels below.[5]
Figure 1: FC Layers
FC Frames
Fiber Channel defines a variable length frame consisting of 36 bytes of overhead and up to 2112 bytes of
payload for a total maximum size of 2148 bytes. A Start of Frame (SOF) delimiter and End of Frame (EOF)
delimiter mark the beginning and end of each Fiber Channel frame.
Figure 2: FC Layers.[7]
Addressing | Naming
World Wide Names
WWN
FC addressing is based upon WWN (World Wide Names). These addresses are unique to each of the FC
devices (such as an FC HBA or SAN's) and are 64-bit addresses, consisting of 16 x Hexadecimal values, like
so: 15:00:00:f0:8c:08:95:de.
Each device in the SAN is identified by a unique WWN. The WWN contains vendor identifier field, which
is defined and maintained by the IEEE.[8]
WWPN
Whereas the WWN is assigned to the device. The World Wide Port Name's (WWPN) is a unique address
assigned to every port of a node (be it HBA or storage system).
WWNN
The World Wide Node Name (WWNN) is assigned to a node in the storage network. For example, the same
WWNN can represent multiple network interfaces of a single network node.[9]
Aliases
Aliases are used to aid in the configuration and troubleshooting, by mapping the World Wide name (from the
FC switch or storage system) to a string. For example: the WWPN 15:00:00:f0:8c:08:95:de could be aliased to
MAIL-SERVER.
Within the FC network, each switch is assigned a unique Domain ID. A principle switch is elected within the
network which assigns the Domain IDs to the other switches. This Domain ID is then used to form part of the
FCID, as per the FLOGI process (described later).
FCNS
The Fiber Channel Name Service (FCNS) is used to exchange the FLOGI database, between switches within
the fabric. Allowing each switch to learn where each WWPN is, and how to route traffic there.[10]
Security
Zoning
Zoning is a security feature that allows the administrator to control which hosts can communicate with each
other. This allows the administrator to create zones that prevent servers from communicating with each other
over the fabric, ensuring that each server can only communicate to the storage system.
Figure 3: Zoning.
LUN Masking
LUN masking is used to restrict which LUN is presented to which host or set of hosts. This prevents
unauthorized hosts from accessing LUNs and also prevents situations of data corruption that could occur
should a host access the wrong LUN.
Ports|Links
Port Types
There are a number of FC port types. Below outlines the main ones:
Port
Port Name Port Description
Type
G_Port Generic This is the port type that all ports first start with, prior to transitioning to another port type.
N_Port Node The Initiator or Target. Connects to an F_Port (aka Fabric switch).
Figure 4: FC Ports.
Link Types
Common Services
Login Processes
The FCID is then used by the FC switches to route traffic across the fabric.
In addition, as part of this process buffer credits are also exchanged with the switch.
WWNN, WWPN
Buffer credits for flow control
Process Login (PLRI)
The initiator host will send a PLRI request to the storage target. The storage system
will then granted access to the host based on its configured LUN masking.
Topologies
Fiber Channel Topologies
Topology is the way of approach to connect devices to form a network.
Point-to-point
Arbitrated loop
Switched fabric
Switched fabric further classified into few more topologies
Traditional topologies
Single switch
Cascading|ring topology
Mesh topology
Tiered topologies
Core edge topology
Edge core edge topology
FC SAN topologies are illustrated in below diagram