Design and Development of IoT based Computer Network Room

Environment Monitoring System

Fahd A. Banakhr

Department of Electrical and Electronic Eng., Yanbu Industrial College, Royal Commission of Yanbu, Saudi Arabia

Keyword: IoT, Internet of Things, NRMS, Network Room, Power, Temperature, Humidity.

Abstract: This paper discusses design and development of internet of thing (IoT) based computer network room

monitoring system (NRMS) for twenty five rooms of college campus. Furthermore, it also discusses the

benefits of Internet of things offer as compare to traditional method of monitoring which does not cope with

rising demand of dynamic power used by servers and networking devices. The NRMS deals with

monitoring network room environmental conditions like power, temperature, and relative humidity. The

system then sends this information to the IoT cloud and displays the live data on the dashboard. It also sends

alert to the authorized persons; records the previous data to find any root cause of the problem if needed.

The data updated periodically from the implemented system can be accessible through internet from

anywhere in the world.

1 INTRODUCTION

The Internet of Things (IoT) is a new growing area

that is gaining ground at very fast pace with virtue of

sophisticated communications technologies. The

basic idea of this concept is the ubiquitous presence

around us of a variety of things or objects various

sensors, actuators, cell phones by using different

protocols and schemes, are able to cooperate with

each other and work together with their neighbors to

reach common goals (Ashrae, 2011). IoT based

NRMS take the benefits of IoT to solve problem and

protect the network to go down. Every network room

is monitored by one NRMS box for multi parameters,

if something goes wrong it take suitable action to

resolve the problem. Thing Speak API is used IoT

base for data recording and root cause analyses of the

problem and serves as an interface between devices

such as temperature, humidity and power sensors to

collect data and analysis software to analyze data to

generate alerts and alarms as required.

2 PROBLEM DEFINATION

The network is room environment is most critical

factor for the efficient working and operating life of

servers and networking switches and other devices,

not proper monitoring can not only damage the

equipment but also halt major work in the

organization.

 The impact of downtime ranges from

direct loss of revenue to the decreased performance

of the organization and spoil reputation. Problem

faced by IT networking devices are listed below.

2.1 High Network Room Temperature

High network room temperature can damage the

hardware of IT equipment present in the room, even

short time rise in the temperature can have worse

consequence. It can create intermittent failure of

equipment as well as permanent data loss. High

temperature can be potential risk for the installation

to catch fire (Stanford University, 2011).

2.2 Low Network Room Temperature

Keeping temperature below the specified limited by

the manufacturer is energy waster, because it require

cooling system to run continuously to maintain the

temperature it can also create health related issues for

the IT personal working in the room. Furthermore it

is heavy on the company budget (ASHRAE, 2008).

2.3 Low Network Room Humidity

Dry climate increases the probability of electrostatic

discharge. Which can cause intermittent or

Banakhr, F.

Design and Development of IoT based Computer Network Room Environment Monitoring System.

DOI: 10.5220/0008925201530160

In Proceedings of the 5th International Conference on Internet of Things, Big Data and Security (IoTBDS 2020), pages 153-160

ISBN: 978-989-758-426-8

153

permanent failure of electronic components.

Repairing damaged components can be expensive

and time consuming (Swenson, D., & J.T. Kinnear,

2009).

2.4 High Network Room Humidity

High humidity level in network rooms can cause

internal components of IT equipment to rust and

change their electrical properties, such as resistance

and capacitance or thermal conductivity. Short-

circuit can be one critical outcome of high humidity,

fire, data storage devices failure and financial loss to

company is consequence (TIA, 2005).

2.5 High Network Room Sound Level

High Sound level in network room is used to monitor

the working of fans of IT equipment, very high sound

level indicate that fan of IT equipment

malfunctioning and to be fixed early as possible to

prevent catastrophic failure. Very High sound in the

server rooms can also cause hearing impairment to IT

personal working or sitting in the room (Dubravko,

2016).

2.6 Low Network Room Sound Level

Low sound level indicate that fan of equipment not

working or running at low speed. Observing sound

levels allows IT personnel to identify a problem as

early as possible.

2.7 Network Room Power Failure

In normal circumstances, Network room power is

backed up by uninterruptable power supply UPS

system some time UPS and main Power failure occur

at same time which halt the communication process

in organization (UPS, 2018).

2.8 Remote Connectivity and Alert

Network room are usually located at isolated place in

the organization, remote monitoring is essential for

these type of system to send alert to IT personal to

take require action.

3 SYSTEM ARCHITECTURE

System level diagram is shown below that shows IoT

cloud service ThingSpeak is used by our system to

display data globally on internet. ThingSpeak is base

platform for the Internet of Things applications.

ThingSpeak allows to build an application around

data collected by sensors on different location. Major

advantage of ThingSpeak include: real-time data

collection, data processing, visualizations, apps, and

plugins. Control room display real time data

collected by different sensors from various location.

Fig.1 shows network rooms are connected with

monitoring systems which is placed in rack or

suitable place connected to internet to take fast and

effective decisions. Data can be displayed on

handheld devices based on android and iPhone. All

the data gathered from network rooms is displayed

control room screen, which gives alerts to the IT

technician if something goes wrong.

Individual sensors do not typically connect

individually to the IP network. Instead, the NRMS

box interpret the sensor data and send data to the

central system by IoT cloud This distributed monitor-

ing architecture dramatically decreases the number of

network drops required and reduces the overall

system cost and management burden. NRMS box are

typically assigned to physical network room within

the organization and combine sensors from a limited

area in order to limit sensor wiring complexity.

ETHERNET ETHERNET

ETHERNETETHERNET

ROOM 1 ROOM 2 ROOM 3 ROOM N

CONTROL ROOM

Monitor Any Where

WIFI/EthernetWIFI/EthernetWIFI/Ethernet

WIFI/EthernetWIFI/Ethernet

WIFI/Ethernet WIFI/Ethernet

IoT Cloud

Figure 1: Network rooms are connected with monitoring

system.

IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security

154

Figure 2: NRMS box connected to one of the servers.

The heart of the IoT NRMS is microchip

PIC32MX360 microcontroller which gives a rich

peripheral set at a low cost for a broad range of

embedded designs that require complex code and

higher feature integration. Block diagram of the

system is shown in figure 3.

The major advantage of PIC32 microcontroller is

one failure every 16.58 years that make it ideal for

monitoring expensive IT equipment. Furthermore, it

can operate at over 100 MHz and can operate in

temperature range from -40 to 105 centigrade. It has

built-in16 channels 10bit ADC that can operate on 1

mega sample per second. Microcontroller has on

chip fail-Safe Clock Monitor that allows safe

shutdown in case of clock failure (Microchip

datasheet, 2011).

PIC32MX360

Error LED

Temperature

Senso r LM35

Multiple

Temperature

Senso rs

DS18B20

An alog

Digital o ut put

On e Wire In pu t

Humidity S en sor

So un d S en sor

I2C

An alog In pu t

Et her n et

WIFI

USB

Power

Monitor

Figure 3: Block diagram of connected system.

3.1 Air Temperature Monitor

LM35 is used to check the air temperature of the

network room The LM35 is low temperature sensor

with an output voltage linearly proportional to the

Centigrade temperature. The LM35 device does need

any external calibration or trimming to provide

typical accuracies of ±0.25°C at room temperature

and 0.75°cover a full −55°C to 150°C temperature

range (Texas instrument datasheet, 2017).

Figure 4: LM35 sensor connected to PIC32 to check the

temperature of the network room.

By using this simple circuit to monitor air

temperature of the network give benefits as written,

 Monitor the room for high temperature due

to cooling system malfunction or total

failure to protect and extent life of

equipment.

 Monitor for very low temperature save

Energy by keeping the temperature at

desired level as consequence save energy

and money.

 Monitor sudden rise of temperature in the

room to protect critical data storage system

from damage. Following equations are used

to calculate the temperature in the software

application.

𝑇𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒 𝑖𝑛 𝐶𝑒𝑛𝑡𝑖𝑔𝑟𝑎𝑑𝑒



LM35 Voltage

10 mV/C

LM35 Voltage  ADC reading ∗

ADC resolution mV/ bit

ADC resolution 

bit

  3.3mV/bit

𝑇𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒 𝑖𝑛 𝐶𝑒𝑛𝑡𝑖𝑔𝑟𝑎𝑑𝑒



3.3 ∗ ADC reading

Design and Development of IoT based Computer Network Room Environment Monitoring System

155

3.2 Network Room Rack Temperature

Monitor

Network Room rack is monitored by using DS18B20

single wire temperature sensors. these sensors can be

network together on singe input pin of

microcontroller with cable length up to 3 meters. The

measurement ranges of these sensor are -50 to

+125⁰C. Each DS18B20 has a unique 64-bit serial

address, which allows multiple DS18B20s to

function on the same1-Wire bus. One digital input is

of pic32 microcontroller is used to control many

DS18B20s distributed over IT equipment rack

(Maxim integrated datasheet, 2015). While

measuring room air temperature in the essential, the

real. concern is the temperature inside the racks.

There are many variables that affect how rack

temperature varies from network room temperature

The only way to monitor for sure that equipment is

operating in a safe temperature is to note the

temperature inside the rack itself. This will save time

in troubleshooting as well as save other equipment

from overheating and extent the overall life.

DS18B20 Temperature = (Temperature High Byte 8

+ Temperature Low Byte) *0.0625

Figure 5 shows single wire connection of four

temperatures sensor placed at different location in the

network room. Figure 6 shows the placement on

temperature sensor at different response (Maxim

integrated, 2015).

DS18B20

DIO

PIC32

Digital

input

3.3

Figure 5: Single wire connection.

3.3 Network Room Rack Humidity

Monitor

The humidity of the air is a measure of the amount of

water vapors content it holds. Relative humidity

(RH) is a convenient way of expressing the amount

of water vapor contained in a volume of air. It's

defined as the ratio expressed in percentage of the

mass of water vapor in the air to the mass required to

DS18B20

NRMS

Box

Figure 6: Placement of temperature sensors.

produce saturation at the same temperature. When

the air is saturated, therefore, its RH is 100%. For

this system low cost Honeywell HIH8120-021-001

humidity sensor is selected, main reason of this

selection is, it can operate on high temperature range

of -40 to 125 centigrade with accuracy of ±2.0 %RH

(Honeywell datasheet, 2015). Moreover, it also

eliminates the need to regularly recalibrate the

sensor, which can be troublesome and pricey. It has

very low power consumption the sensor goes into

sleep mode when not taking Measurement,

consuming only 1 μA versus 650 μA figure 7 shows

hardware interface to humidity sensor with

microcontroller by simple I2C protocol. This solution

protects equipment failure from static electricity

accumulation at low humidity points and

condensation formation at high humidity points on

the IT equipment. Following formula is used to

calculate the humidity (%RH).

Humidity %RH



Humidity Output Count ∗ 100

16383

IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security

156

HIH8120-021-001

VDD SCL

SDAVSS

3.3

PIC32

Figure 7: Hardware interface to humidity sensor with

microcontroller.

3.4 Network Room Sound Monitor

Sound monitor circuit uses SEN-14262 sound

detector based on microphone and audio amplifier

(Spark Fun online, 2018). The Sound monitor has

three separate outputs. One is amplified audio output

from microphone, second is sound envelope and third

is pulse output when sound above threshold is

detected. Microcontroller analog pin is connected

with envelope pin of sound detector to identify audio

level which indicate malfunctioning of fans

Sound Detector

Read Analog Value from ADC

If sound level is greater than 30 means very

noisy,

may be fans are dusty.

If sound level is less than 10 fan is not

working

check the fan

3.5 Network Room Power Monitor

Simple opt coupler is used to detect the power failure

from the main socket this simple circuit work on

UPS power to detect the condition of main power. If

is power for long time it will send alert to concerned

maintenance technicians.

Ethernet cable of NRMS is connected with main

switch or hub in when power is loss by the switch

communication loss error due to power is send to the

control room.

PIC32

3.3VDC

Figure 8: Power failure detection.

3.6 NRMS Wifi Communication

In this project ESP8266 IoT wifi module is used

which is very low cost with a complete AT command

library. This allows for stress-free integration with a

Wi-Fi network through serial communication.

Module has low power consumption at sleep mode

(Spark Fun online, 2018).

PIC32

VCC

Figure 9: ESP8266 IoT WiFi module.

4 SYSTEM SOFTWARE

DESCRIPTION

System software has four major components

Application programming interface API and Web

Service for the Internet of Things, embedded

software for microcontroller, LabVIEW program for

main control panel software, and mobile software for

remote monitoring. Following paragraphs discuss the

implementation.

4.1 API and Web Service for the IoT

For the purpose of connecting an NRMS to the IoT,

system uses ThingSpeak API (ThingSpeak, 2018).

Fig 10 represents the connections of system

The API provides simple communication capabilities

to objects within the IoT environment, as well as

Design and Development of IoT based Computer Network Room Environment Monitoring System

157

interesting additional feature like tweeter interface.

Additionally, ThingSpeak permits to build software

applications around data collected by sensors by

NRMS Data is stored in. the channel that is primary

component of ThingSpeak, which holds data fields,

location fields, and a status field. Real-time snapshot

of thingspeak is shown in the figure 11.

Create a Channel and

collect data

Analyze and Visualize

the data

Act on the data

PIC32

Sensor

ThingSpeak

Temperature Room 1

Temperature Room 2

Humidity Room 1

Humidity Room 2

WIFI

Module

Webpage

Sensor

Senor

Figure 10: Connection system of NRMS to IoT.

Figure 11: Real-time snapshot of thingspeak.

4.2 PIC32 Microcontroller Ethernet

Program

The microcontroller uses microchip TCP/IP stack

that provide a foundation for embedded IoT network

applications by handling most of the interaction

required between the physical network port and

sensing application. They stack provide for

complicated network layers free of cost on embedded

microcontroller that include HTTP for serving web

pages, SMTP for sending e-mails, SNMP for

providing status and control. Microchip’s TCP/IP

stack uses of cooperative multitasking to implement

the various TCP/IP stack functions and collect sensor

data, which all cooperate, to share processing time on

a single microcontroller. In other words,. This is done

by either dividing its job into multiple tasks, or

organizing its main job into a Finite State Machine

(FSM) and dividing a long job into multiple smaller

jobs. Each job has been planned to cooperate by

running only a short time so that the other tasks can

have their share of the processor. State diagram of

microcontroller program is shown in fig. 12.

Connection

Established

Disconnected

Read sensors

data

Initial

Failed

Success

Yes

Quit

Open Socket

Disconnect

Not finish

Figure 12: State diagram of microcontroller program.

4.3 PIC32 Microcontroller Wi-Fi

Program

The ESP8266 Wi-Fi module has a full TCP/IP stack

support. It can effortlessly have configured as a web

server or client by use of serial port of

microcontroller. The module accepts serial AT

commands and responds back with the operation's

outcome. Moreover, once the device is connected

and is set to accept connections, it will send

spontaneous messages whenever a new connection or

a new request is issued. Communication algorithm

IoT cloud and Wi-Fi module is shown in figure 13.

IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security

158

START

Reset Module

AT+RST

Configure Access point

AT+CWMODE

Ge t IP add ress

AT+CIFSR

Multiple Connections

AT+CIPMUX=1

Read temperature from

LM35

Read temperature from

DS18B20

Read Relative Humidity

Read Sound data

Read Power status

Establish TCP Connection

AT+CIPSTART

Send Sensor Data to

IoT Cloud

AT+CIPSEND

Establish Serial

connection with ESP

END

Figure 13: Communication algorithm IoT cloud and Wi-Fi

module.

4.4 Main Control Room Software

Main control software display information from the

all network rooms in one place developed using

LabVIEW (LabVIEW, 2017). Figure 14 shows

sensors and power information displayed on main

screen on the control room.

Figure 14: Sensors and power information displayed on

main screen on the control room.

User can set the desired threshold of sensors for

generating alerts and alarm from main control panel

software for IT Staff. Based on the severity level of

an alert system decide the action to be taken These

automated actions could be personnel email or cell

phone text message notification, or they could be

corrective actions such as triggering back up cooling

fans or dryer

Historical recoding of sensors data is implemented in

the program Data can be used to predict future

behavior as more and more equipment are aggregated

in the room, and could help forecast when the data

room will reach to the limit. Long term historical

data analysis can be used at the rack level to compare

how equipment from different manufacturers in

different racks, which may guide future purchases

recording.

5 RESULTS

Table 1 shows the result of Implementing the NRMS

in different locations in the campus.

Table 1: RCYCI Communication Room Temperature.

Location

MAC

Address

BOX

Mean

Temp

Mean

Relt.

Humidity

Power

Status

Sound

Level

YIC Main

Data

Center

00-04-

A3-F4-

C0-06

NRMS0

20 62 OK OK

YIC ACX

Comm

room

00-04-

A3-F4-

BB-04

NRMS0

22 60 OK OK

YIC AC

Server

Room

00-04-

A3-F4-

C0-1B

NRMS0

21 61 OK OK

YIC New

Dorm B#1

00-04-

A3-F4-

BB-46

NRMS1

25 60 OK OK

YIC New

Dorm B#7

00-04-

A3-F4-

B7-F0

NRMS0

24 63 OK OK

YTI Main

Server

Room

00-04-

A3-F4-

BC-97

NRMS0

20 62 OK OK

Design and Development of IoT based Computer Network Room Environment Monitoring System

159

Table 1: RCYCI Communication Room Temperature

(cont.).

Location

MAC

Address

BOX

Mean

Temp

Mean

Relt.

Humidity

Power

Status

Sound

Level

YTI B2

Comm

Room

00-04-

A3-F4-

BE-F5

NRMS1

21 60 OK OK

YTI B4

Comm

Room

00-04-

A3-F4-

BE-99

NRMS1

26 59 OK OK

YTI B5

Comm

Room

00-04-

A3-F4-

C8-18

NRMS0

58 OK OK

YUCM J1

Server

room

00-04-

A3-F4-

B7-20

NRMS1

20 60 OK OK

YUCM J2

Server

room

00-04-

A3-F4-

BF-A9

NRMS0

23 61 OK OK

YUCM J1

Comm

room

00-04-

A3-F4-

BB-65

NRMS1

24 62 OK OK

YUCF

Buildg A

00-04-

A3-F4-

C0-4C

NRMS1

61 OK OK

YUCF

Buildg B

00-04-

A3-F4-

A5-04

NRMS0

20 60 OK OK

YUCF

Buildg C

00-04-

A3-F4-

C2-40

NRMS1

22 62 OK OK

YUCF

Buildg D

00-04-

A3-F4-

BB-0D

NRMS1

21 63 OK OK

YUCF

Buildg E

00-04-

A3-F4-

C0-45

NRMS1

23 64 OK OK

Monitoring system

*YIC, refers to Yanbu Industrial College.

*YUCM, refers to Yanbu University College Male

Campus.

*YUCF, refers to Yanbu University College Female

campus.

*YTI, refers to Yanbu Technology Institute.

6 CONCLUSION

IoT based NMRS provide protection against

environmental threats that is critical to a

comprehensive network equipment safety strategy.

Location and procedure of temperature, humidity,

and sound sensing equipment requires assessment,

choice, and design, best practices and design tools

are available to support for efficient and effective

deployment. IoT software and control program

manage the collected data and provide recording,

analysis, smart alerts and automated corrective and

preventive action as desired. By utilizing IoT for

monitoring dispersed physical threats in different

network room enables the IT administrator to fill

critical cracks in overall network safety to achieve

the required goals.

REFERENCES

ASHRAE, 2011. Thermal guidelines for data processing

environments – expanded data center classes and

usage guidance. TC 9.9, ASHRAE, Atlanta, USA,

2011.

Stanford University Server/Telecom Rooms Design

Guide. Stanford, CA: Stanford University, 2011.

ASHRAE. Environmental Guidelines for Datacom

Equipment— Expanding the Recommended

Environmental Envelope. Developed by ASHRAE TC

9.9. 2008.

Swenson, D., J.T. Kinnear. 2009. The Role of Relative

Humidity and Dew Point on Electrostatic Charge

Generation and Discharge (ESD). The Green Grid.

TIA-942-2005. Telecommunications infrastructure

standard for data centers, Washington, DC: American

National Standards Institute, 2005.

Dubravko,”MiljkovićNoise within a data center”, In Proc.

Information and Communication Technology,

Electronics and Microelectronics (MIPRO), 2016.

UPS failures continue to be the top cause of data center

downtime [online] http://www.geistglobal.com/news/

company-blog/ups-failures-continue-be-top-cause-

data-center-downtime [Accessed jan. 10, 2018]

Microchip “PIC32MX3XX/4XX Family Data Sheet”,

2011.

Texas instrument “LM35 Precision Centigrade

Temperature Sensors”, Datasheet 1999, [Revised

2017].

Maxim integrated “ DS DS18B20 Programmable

Resolution 1-Wire Digital Thermometer”, datasheet,

2007 [Revised 2015].

Maxim integrated ,Guidelines for Reliable Long Line 1-

Wire Networks”, Application notes 148.

Honeywell “HumidIcon™ Digital Humidity/Temperature

Sensors, HIH8000 Series”, Datasheet, 2015.

Spark Fun Sound Detector “https://www.sparkfun.com/

products/14262 “[Accessed jan. 10, 2018].

WiFi Module - ESP8266,”https://www.sparkfun.com/

products/13678”, “[Accessed Jan. 9, 2018].

ThingSpeak,” https://thingspeak.com/”, [Accessed Jan.11,

2018].

LabVIEW , LabVIEW 2017, http://www.ni.com/”.

IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security

160