Building an Array Utilizing RAID Setup

This document discusses the process of building an array utilizing RAID setup, the different RAID levels, and the steps to configure RAID on a computer. It also provides information on designing a system with 4 x identical 1TB hard disks.

NETWORKING

2
NETWORKING
Table of Contents
Lesson 9...........................................................................................................................................3
Building an array utilizing RAID setup.......................................................................................3
How you design a system with 4 x identical 1TB hard disks......................................................6
Lesson 10.......................................................................................................................................10
Design considerations................................................................................................................10
Handle the USB devices............................................................................................................13
Reference list.................................................................................................................................15
Appendices....................................................................................................................................19

3
NETWORKING
Lesson 9
Building an array utilizing RAID setup
RAID (redundant array of independent disk) is the process of storing the same data on
multiple hard drives to protect from the failure of hard disks. All RAID levels do not represent
redundancy. The system works by putting data on multiple drives and start input-output
operations to overrun a balanced path. The use of different drives helps to shorten the time
between failures; this will also grow fault tolerance.
RAID arrays are introduced to the computer operating system as an individual logical
hard drive. The main function of this drive is to copy mirror data on different hard drives
available. The system separates each partition from 512 bytes to a lot of megabytes (Kreutz, et
al., 2015). Stripes of the disks are interlaced and placed in order. The size of RAID depends upon
user systems. Single user systems require a small number of data storage, but in an organization,
a lot of data is stored in hard disks.
Disk mirroring can be done using together in RAID 1 and RAID 10. A RAID controller
is used as a detachment between the operating system and hard drives. The controller is used for
improving performance and helps to protect data from crashing. This can be joined or designed
as a support drives like SATA and SCSI, it can also be a component of the motherboard.

4
NETWORKING
Fig 1: Different RAID Levels
(Source: Created by Author)
Different types of RAID levels are available for users like RAID adapters, Windows
software RAID, motherboard chipsets and consumer level (Maier, Laumer, Eckhardt & Weitzel,
2015)
JBOD is the type of RAID level which is used to extend the volume on other drives. The
level has no development in performance and it is useless nowadays to handle a large amount of
data. RAID 0 is used for disk stripping and crossing data between two drives. Dividing data into
two data busses this can be read or write easily. This level does not provide data protection and
increases the possibility of crashing failure.
RAID 1 provides full data protection between different data on hard drives. This level
reduces the chance of failure by 50%. Provide a 2TB drive for 2TB storage. RAID 1 generally
used for disk mirroring. RAID 5 requires at least three hard drives. The system distributes

5
NETWORKING
redundancy information instead of using a single hard disk. A replacement drive is also added in
the system. This level generates only 33% of available storage. RAID 10 is a combined version
of RAID 1 and RAID 0. The level works by striping data around two devices or stripping data
between two mirrored pairs. This level requires a minimum of four hard drives (Parker, 2016).
A fresh backup is required before installing RAID on the computer. Configuring
hardware RAID on a new system with no operating system adds different disk for a customized
array. Backup of data and make a copy at least three times is more preferable. Before creating a
new array user must back up the previously existing data in the new hard drives (Pathak, Feng,
Hu & Mohapatra, 2015).
The process writes the new data into a hard drive. This is not crucial for Windows
software RAID 0 and it is faster than Windows re-sync. Windows Software RAID was started
with Microsoft XP, the software allowed JBOD. Windows vista grants both JBOD and RAID 1
striping. Windows 7 grants JBOD and RAID 1. This type of RAID gives different advantages.
Create an array from Windows Drive Manager and no BIOS setting is required to configure and
it is also flexible with mirrors.
The mirrors help to create previously existing data and delete half data with the
remaining. The user can mirror different partitions, adding partitions from particular drives into a
single drive. Disadvantages of the system are operating systems such as Linux cannot able to see
RAID partitions. Nowadays Windows RAID is generally fast, the restoring mirror is not simple
with a hardware problem-solving issue. The secondary RAID controller is included inside the
motherboard. New motherboard chipsets include integrated RAID and interfere with an
additional controller. Check the motherboard documents carefully to find out if the motherboard
supports RAID (Shang et al., 2016).
The user can use an adapter card, Basic PCI and PCIe to enable the RAID levels 0,1,10
and level 5. A card that supports more than 2.2TB disk and 6GB per second SATA III is the best
buyer option for a user. Access the Disk Utility tab through application utilities then go to the
chosen drives to generate RAID 1. Click on RAID 1 and enter a RAID set name to rename the
drive. Choose the mirrored RAID set. Go to disk management tool right click on the unallocated

6
NETWORKING
space on the dynamic disk. Create RAID 5 volume and after that choose Create Volume. Click
F2 during startup to access the BIOS setup. Go to the configuration then choose SATA drives
and change the Chipset SATA mode to RAID. Configure RAID in Windows 10, search ‘Storage
Spaces' in the search tab choose to create a new pool select RAID type under the drop down
menu and then choose to create storage space (Xia, Liu, Li, Ma & Vasilakos, 2015).
How you design a system with 4 x identical 1TB hard disks
Hard drive or often called disk drive is storage used in computers or laptops that provide
storing data in a large amount (Blenk, Basta, Reisslein & Kellerer, 2016). It uses the charged
surface of electromagnetic. Nowadays computers come with large capacity hard disks. A hard
disk is similar to stacked disk such as phonograph records. Each data is recorded
electromagnetically using concentric circles. Two heads are there on either side of the disk and
they record data when the disk spins. Read or write operation is required in this system to locate
data. Hard drive units come with different rotation speed ranges from 4200 rpm to 15,000 rpm.
Most of the computers and laptops come with 5400 rpm to 7200 rpm (Burusic & Karabegovic,
2016). However, hard disk with high rpm is used in workstations. The access time of disk is
measured in milliseconds. Apart from that, cylinder, track and sector locations are used in the
disk drive to map the logical block address.
The hard drive is still popular with enterprises and consumers. Lately, another type of
hard drive has gained popularity and that is called SSD or solid state drive. It is actually a flash
drive with a non-volatile solid state. SSD also fits in the computer in the same counterpart. The
reason behind the growing popularity of SSD is it relatively faster than HDD. It is also more
powerful than and durable than HDD but it is more expensive and produces less storage. The
hard disk was first created by IBM engineers and since then it was no looking back. However,
the size of the hard disk used to be large but in less storage. Later, Memorex implemented new
technology and made IBM compatible hard disk. Since then, Western Digital and Seagate came
in this field and introduced much faster HDDs and SSDs. The first patent used a redundant array
of the independent disk or RAID technology.
Hard disks come with ranges of capacity and in recent days capacity has been developed
with improved speed. Seagate, Western Digital is some of the famous companies in the disk

7
NETWORKING
drive sector. Building and designing capacity require some stages (Engel, Kaandorp & Elfring,
2017). Earlier hard disks were filled with helium. Helium is helpful because of less dragging and
turbulence because. It is also helpful too because of less dense and light air. Helium hard disks
are manufactured to use seven platters in the same space which is required in conventional hard
drives. Lately, Seagate has introduced shingled magnetic recording or SMR technology in hard
drives. It has gained popularity as it became helpful to overcome limitations in conventional
drives. SMR technology is layers of magnetic tracks used in each disk in a parallel way in
conventional hard disks. The technology is about overlapping the shingles.
Fig 2: RAID technology in hard disk
(Source: Created by author)
Several parts are there in a hard drive and those are data storing platter, spinning platter
spindle, arm for reading and writing, an actuator etc (Haleplidis et al., 2015). Main works for
actuator are controlling movement and action of reading and write arm with the logic board.
Hard disks consist of a number of things such as aluminum, the ceramic platter of substrate
material using thing magnetic surface or layers of media to store data. The work of the platter is

8
NETWORKING
to store data in an organized and specific way. This includes tracks, clusters, and sectors. The
hard drive also has a lubricating protective layer and a superthin protective layer in order to
avoid accidental damage and dust. The spindle makes the platter to rotate and helps to hold in
position. Reading or writing data is determined with how the rpm is. Multiple platter HDDs
spindle maintains a fixed platter with a separate distance to make room for operating. The disk
platter is needed to control arm position of reading and write. Generally, there is a floating of 3
to 20 millionth of a bottom and top inch on either surface of every platter (Hwang,
Ramakrishnan & Wood, 2015) The actuator motor is fused together to read and write of the arm
of the disk. HDD circuit board controls and gives instruction to actuator motor. An intelligent
circuit board is used which gives instruction to actuator motor for what needs to do. It places at
the bottom base of the unit. It is known as casing and cable connection of flexible ribbon which
connects the circuit board with the actuator.
Hard disk follows RAID technology or redundant array of the independent disk. It is a
way of storing data in different places of the hard disk. It also helps to protect data in case of
drive failure. RAID works by placing multiple data on multiple disks which allows input (I/O) to
balance overlapping. More than two hard drives are configured with RAID technology and it
becomes helpful to improve performance and drive failure loss. Earlier RAID was expensive but
the rapid implementation of this technology has been helpful to reduce cost. Most of the
motherboards in moderns desktops and laptops support RAID technology. Even, new operating
systems also support RAID technology. Now, it is needed to remember what kind of RAID is
required to set up on the computer. RAID has several layers and the most used layers range from
RAID 0 to RAID 6. Some RAID levels are described below to design and set up on PC.
JBOD (extend, spill over) It refers to bunch which allows the volume to extend. Data is
usually written in the first disk and it transfers to other parts when the drive is full. Then it
transfers to the third part and so on. It does not offer to boost in performance when the disk is
loaded with the full amount of data.
RAID 0 (Stripping) this method is helpful because it increases the power of hard drive
by splitting. Sometimes it is known as stripping also. Data leveraging in different locations
becomes helpful to read and access data more quickly. Unfortunately, RAID 0 does not provide

9
NETWORKING
any data protection. Moreover, it creates a chance to lose data. This method is used in
workstations and gaming personal computers.
RAID 1 (Mirroring) This method helps to set and protect data from failure. It is done by
using writing the same data simultaneously on two hard drives. It is an exact duplicate of other
drives and it helps to restore data if one part fails.
RAID 5 (Distributed Parity) this part helps to spread additional information. It requires
33% storage whereas RAID 1 requires 50%.
RAID is needed to set up a fresh back up in existing data only with the compatible OS.
Setting up an array of the disk can be divided into a few parts. One of them is Windows
software red. This is done by using features of windows software. Windows XP does allow
spanned volume aka JBOD. Windows Vista helps to allow JBOD and RAID 1 striping. Windows
7 allows JBOD and RAID 1 but professional provides RAID 0 mirroring. On the other hand, the
windows server uses JBOD, 0, 1 plus with distributed parity (pubsonline.informs.org, 2019).
RAID controller is needed to use in a personal computer in order to improve the
performance of the hard drive. Older and midrange motherboards support RAID technology but
it requires installation when a controller is missing. Other steps are required to maintain when
there are two or more hard drives. They may not be in the same size if individual volumes are
mirrored. Earlier it used to ATA in RAID controller but now it uses SATA technology which is
faster and prevents data loss. Controllers are there to mix and match capacities without loss of
space storage.

10
NETWORKING
Lesson 10
Design considerations
A new I/O device is developed for use, where the device will have a touch screen display
and must consist of a microphone, webcam and also USB plugs to attach the device with other
ones.
During making the design of an I/O device it is necessary to know clearly the
requirements of the device first. The main requirement of the I/O system is about satisfying the
latency of the device. Fulfilling this requirement requires the critically timing operation. It is
necessary to design a device that can be used as less time consuming (Khan, Rehmani &
Reisslein, 2016).
Operations that are needed to be performed in this case must be critical time managed. In
cases where the system is unloaded, the design must be done as latency will be added up in
components (Vacca, 2015).
Besides, the device needs to be maximized throughput. In this work, the first thing a
developer needs to find out is the weakest link of the system. The component that has the lowest
bandwidth will be considered as the weakest link of the device. Then the design must be changed
and configure the link as making its higher bandwidth. Other components of the device are kept
in balance during the changing in the bandwidth of components. In other cases where the system
is loaded, the models are needed to be kept in a queue or making it simulate. This is also done
when the analysis of the components is not that simple. The weakest links of any I/O device is
noticed to be the disks and memory spaces used in it. In random I/O rate of the disk is found to
be 2424 ops/sec. The considerations which are necessary for designing the I/O device are:
● Listing the standard buses that will support the device. [Referred to Appendix 1]
● Listing the physical requirements of the device, like connectors, expansion cabinets, bus
slots, size and power and so on.

11
NETWORKING
● Listing the actual cost of the device costs and the costs of the components used in it.
● Listing the CPU resources requirements for the device, like the clock cycle to initiate the
work of the device, CPU stalls, cache flush, and many others.
● Listing the memory and bus resource requirements of the I/O device. It is needed to be
kept in mind that the bandwidth of memory remains limited even if the CPU is not in
running condition.
● Assessing the device performance and also making sure that there are different methods
in overcoming the issues if occurs in I/O device in the future. [Referred to Appendix 2]
The design considerations which are required for making a touch screen display of I/O device
are:
● Providing enough space for the users as there will no issue of touching other tabs due to
lack of space. Mobiles, Monitors are changed in recent days and increased in size for this
purpose. This is achieved according to the customer's demands.
● Providing fewer options and by this, there will be less confusion in the users' mind in
designing a touch screen device. Limited windows, limited boxes need to be kept in the
device for reducing confusion (arxiv.org, 2019).
The design considerations for making the I/O device usable of microphones and web cameras
can be said as:
● Providing enough space and a port for entering the microphone plug within it. Using
proper receiver components in the plug for supporting every kind of microphone.
● Providing a web camera in front of the monitor which will be helpful to make video
calling and video recording.
● The sound receiver component must be entered in the device, which will gather the voice
messages during video recording.
Communication tools used (500)
There are different types of communication methods used in the I/O device for making
proper communication with all the components to the CPU. It can be classified into three ways
for communicating with CPU to other components:

12
NETWORKING
● I/O Memory mapped: During this process, similar address space is shared with I/O
device and the memory. In this case, the device is directly connected to the main memory
source. This will help the device to transfer the data to the memory and from the memory.
The data from memory is then transferred to or from the CPU.
During this process, memory allocates a buffer within it and then allows the I/o device to
use that buffer to send and collect data to or from the CPU. These activities are done by
the device asynchronously (operating concurrently with the CPU activities). The device
also tends to interrupt the CPU when actions are finished (ece.mtu.edu, 2019).
● I/O Special instruction: The instructions are made basically by the CPU and sent to the
I/O device as per the requirement of the work. Instructions work as a controlling device.
● DMA (Direct memory access): There are many scenarios where the work of CPU is
interrupted when a single bit is transferred. Slow devices occur such problems in the
device any time. This will happen in keyboard case, but for the fast devices like disk
makes interruption in CPU work, the computer takes more time in handling this. For this
scenario, DMA is used for reducing the workload of the device.
DMA allows the I/O device to access the memory, read from it or write on it as per the
requirement. DMA helps in transferring data between the I/o device and the memory.
CPU is not involved in this task, except in the beginning and end of the activity. When
the entire information packet is transferred, the CPU gets into the work.

13
NETWORKING
Fig 3: Communication tools for I/O device
(Source: Created by author)
DMA also requires hardware (DMA controller) for accessing the system bus. These
controllers are programmed with the sources (static.usenix.org, 2019).
The communication tools that are needed for the components of the device are I/O ports,
LCDs such as VGA, DVI, video capture cards, sound cards, and many others. USB cables are
inserted into USB ports. For making a better communication and transfer of information to the
CPU and devices, input ports must be selected carefully. USB port above than 1.1 will be helpful
for making easy access to data and it will also provide high speed in data transferring.
Handle the USB devices
Some of the USB devices like printers, scanners, and copiers required USB driver
software before installation in a computer. The power source of the computer must be stable
cause USB devices depend on the power sources. Make sure that all connections are secure and
the power cable of the computer connected to a proper power source. Sometimes reconnecting
the USB device solves the problem. Disconnect all the devices from the computer except mouse
and keyboard. System requirement is also a notable thing. The latest versions of operating
systems and the installation of new hardware will easily load the USB devices. Microsoft Easy

14
NETWORKING
Fix Solutions is a web-based application which is generally helpful to find the problem and give
proper solution to the issue. The older versions of operating systems do not support the latest
devices. Make sure to update to the latest version of the operating system. Problems are created
when the installation of a new USB device if the device driver is already installed on the
computer then it will detect the hard drive fast (patents.google.com, 2019).
The USB cords are connected through connectors, the connectors can be of two types
series A and series B. In these types series A is used on the USB hubs and helps to support the
USB devices and in other case series, B is used for the device which employs removable USB
cable. There are various types of USB ports that can be used in the I/O devices. USB 3.0 is the
modified version that can be used here as it gives a speed of around 4800Mbps. USB 1.1 is cost-
effective to use in the device (ieeexplore.ieee.org, 2019).
The device required a USB port for the microphone as the device will be used for recording
videos. There must be multiple ports available in the device which will allow connecting many
devices and systems with it in the future. The methods of connecting USB ports can be done by
following methods:
● Add on cards
● Hubs
● Motherboard connectors

15
NETWORKING
Fig 4: Methods of connecting USB ports
(Source: created by author)
Bus powered hubs will be useful for designing this device and plugging USB devices in it.
USB devices plugged into the display must be handled with proper methods. The touch screen is
used in the display. So it will be easy to handle the USB devices that are connected in the
display.

16
NETWORKING
Reference list
Book
Vacca, J. R. (2015). Handbook of sensor networking: advanced technologies and applications.
Chapman and Hall/CRC, Retrieved from
http://dspace.fue.edu.eg/xmlui/bitstream/handle/123456789/3672/10541.pdf?
sequence=1&isAllowed=y
Journals
Blenk, A., Basta, A., Reisslein, M., & Kellerer, W. (2016). Survey on network virtualization
hypervisors for software-defined networking. arXiv preprint arXiv:1506.07275.
Burusic, J., & Karabegovic, M. (2016). The Role of Students' Personality Traits in the Effective
Use of Social Networking Sites in the Educational Context. In Psychology and Mental
Health: Concepts, Methodologies, Tools, and Applications (pp. 1205-1224). IGI Global.
Engel, Y., Kaandorp, M., & Elfring, T. (2017). Toward a dynamic process model of
entrepreneurial networking under uncertainty. Journal of Business Venturing, 32(1), 35-
51.
Haleplidis, E., Pentikousis, K., Denazis, S., Salim, J. H., Meyer, D., & Koufopavlou, O.
(2015). Software-defined networking (SDN): Layers and architecture terminology (No.
RFC 7426). Retrieved from: http://www.rfc-editor.org/rfc/pdfrfc/rfc7426.txt.pdf
Hwang, J., Ramakrishnan, K. K., & Wood, T. (2015). NetVM: high performance and flexible
networking using virtualization on commodity platforms. IEEE Transactions on Network
and Service Management, 12(1), 34-47.
Khan, Athar Ali, Mubashir Husain Rehmani, & Martin Reisslein. "Cognitive radio for smart
grids: Survey of architectures, spectrum sensing mechanisms, and networking
protocols." IEEE Communications Surveys & Tutorials 18, no. 1 (2016): 860-898.

17
NETWORKING
Kreutz, D., Ramos, F. M., Verissimo, P. E., Rothenberg, C. E., Azodolmolky, S., & Uhlig, S.
(2015). Software-defined networking: A comprehensive survey. Proceedings of the
IEEE, 103(1), 14-76. Retrieved from: https://arxiv.org/pdf/1406.0440.pdf
Maier, C., Laumer, S., Eckhardt, A., & Weitzel, T. (2015). Giving too much social support:
social overload on social networking sites. European Journal of Information
Systems, 24(5), 447-464.
Parker, B. J. (2016). U.S. Patent No. 9,311,160. Washington, DC: U.S. Patent and Trademark
Office.
Pathak, P. H., Feng, X., Hu, P., & Mohapatra, P. (2015). Visible light communication,
networking, and sensing: A survey, potential, and challenges. IEEE communications
surveys & tutorials, 17(4), 2047-2077.
Shang, W., Bannis, A., Liang, T., Wang, Z., Yu, Y., Afanasyev, A., ... & Zhang, L. (2016,
April). Named data networking of things. In Internet-of-Things Design and
Implementation (IoTDI), 2016 IEEE First International Conference on (pp. 117-128).
IEEE.
Xia, F., Liu, L., Li, J., Ma, J., & Vasilakos, A. V. (2015). Socially aware networking: A
survey. IEEE Systems Journal, 9(3), 904-921.
Online Articles
arxiv.org/ (2019), Leaking (a lot of) Data from Air-Gapped Computers via the (small) Hard Drive
LED, retrieved from https://arxiv.org/ftp/arxiv/papers/1702/1702.06715.pdf [Retreived on
28.01.2019]
ece.mtu.edu (2019), Solid State Drives Data Reliability and Lifetime, retrieved from
http://www.ece.mtu.edu/faculty/rmkieckh/cla/4173/REFERENCES/SSD-Imation-WP.pdf
[Retrieved on 28.01.2019]

18
NETWORKING
static.usenix.org (2019), DFS: A File System for Virtualized Flash Storage, retrieved from
http://static.usenix.org/legacy/events/fast10/tech/full_papers/fast10proceedings.pdf#page
=93 [Retrieved on 28.01.2019]
Websites
ieeexplore.ieee.org (2019), Hard disk drive servo technology for media-level servo track writing
retrieved from https://ieeexplore.ieee.org/abstract/document/1190113 [Retrieved on
28.01.2019]
patents.google.com (2019), Hard disk drive apparatus, method to compensate repeatable run
out of hard disk drive apparatus, and computer-readable recording media recording the
method, retrieved from https://patents.google.com/patent/US20080175113A1/en
[Retrieved on 28.01.2019]
pubsonline.informs.org (2019), Past Success and Creativity over Time: A Study of Inventors in
the Hard Disk Drive Industry retrieved from
https://pubsonline.informs.org/doi/abs/10.1287/mnsc.1060.0593 [Retrieved on
28.01.2019]

19
NETWORKING
Appendices
Appendix 1: I/O bus connection
(Source: https://slideplayer.com/slide/12469035/)

20
NETWORKING
Appendix 2: I/O system organization example
(Source: http://www.ecs.umass.edu/ece/koren/architecture/IO/intro.html)