Server Availability Monitoring and Metrics

This research report compares and evaluates different server availability monitoring tools and their metrics. It discusses the importance of monitoring server availability and suggests using VPN and service auditing as solutions. The report also highlights the benefits of failure prediction and concludes with the recommendation of Nagios and Zabbix as the best tools for server monitoring.

Surname 1
Server Availability Monitoring and Metrics
Student Name
Student Number
Professor’s Name
Institution
Location of Institution
Date
Table of Contents
1 Introduction..............................................................................................................................................3
2 Comparison..........................................................................................................................................of 3
2.1 Disk Usage.........................................................................................................................................3
2.2 CPU Monitoring.................................................................................................................................4

Surname 2
2.3 Process Monitoring...........................................................................................................................5
2.4 Bandwidth Monitoring......................................................................................................................6
2.5 Resource Measuring..........................................................................................................................7
2.6 Overall Availability.............................................................................................................................8
3.1 Nagios................................................................................................................................................9
3.2 Zabbix..............................................................................................................................................10
4 Solution proposal....................................................................................................................................10
4.1 VPN and the Private Networking.....................................................................................................10
4.2 Service Auditing...............................................................................................................................10
5 Failure Prediction Impact........................................................................................................................11
6 Conclusion..............................................................................................................................................11
References.................................................................................................................................................12

Surname 3
1 Introduction
Security is one of the factors to be considered in every organization. In all the operations, data
management is purely done in the server side and this calls for the management and the overall
monitoring of the working of the overall system [1]. Since most of the operations run on a full-time
basis, no one can be able to see all corners of the organization at an ago.
To evade from system server failure, there is a need to have some tools that guarantee proper
monitoring. In this research report, we are going to compare the performance and the utility of some of
the server availability monitoring tools including but not limited to; Nagios Core, Ganglia, CollectD,
Spiceworks, Cacti, Zabbix, Icinga, OpenNMS and WhatsUp Gold with respect to the metrics we will
consider in the next section.
2 Comparison
2.1 Disk Usage
Nagios Core Ganglia CollectD Spiceworks Cacti
This uses the disk
well in terms of
the size unlike
percentage, this
tool uses the
least space
possible
Uses each of the nodes
within the disk cluster
to either multicast the
state of the host or the
XML transmission over
the connection.
It uses RRD files to
store the metrics
systematically which
can be retrieved
periodically ate the
specified time.
Uses most of the
disk storage
while it can as
well be installed
on a virtual
machine for
other resource
performance.
Uses the RRD tools
for all its data
storage. Storage is
based on a compact
mode where no
extension is allowing
to expand over a
given time.
Zabbix Icinga OpenNMS WhatsUp Gold
XML based storage. It
uses this configuration
for optimum
Uses little storage
space, unlike other
Just like a website, it
uses a small storage
space. The Up and
Requires external disk
storage hence it can
exhaust the whole disk

Surname 4
consumption of the
disk
management tools. Down devices allow for
a small disk space to be
used.
space
2.2 CPU Monitoring
Nagios Core Ganglia CollectD Spiceworks Cacti
It provides a
well-utilized CPU
utilization
Ganglia does not
guarantee CPU
speed in maximum
performance
Shares the
information
collected for the
performance in
the form of the
auxiliary
functions.
Allows addition of
the memory and
the CPU to the list
of the monitored
resources.
With the
intervals in the
resultant data,
the graph
series of time
allows proper
data traffic.
Zabbix Icinga OpenNMS WhatsUp Gold
Uses an agent to
monitor the host and
the related operations.
Monitors both the
remote and the local
resources.
Uses SNMP for
optimum monitoring
which needs to be
configured first before
being deployed.
Displays usage in a
form of graph and is
always on a minimal
scale.no much straining
in the usage.
2.3 Process Monitoring
Nagios Core Ganglia CollectD Spiceworks Cacti

Surname 5
Provides total
monitoring and
the respective
states.
Has a web
interface tabs that
monitors
processes
independently
and easy
navigation to the
tab you need
directly from the
dashboard.
It gathers all the
data regarding the
process that runs
on the host
computer. It
automatically
switches between
the stated
processes.
It has an offline
feature that
monitors the
processes in case of
a fall in network
coverage.
With cacti,
process
monitoring is
allowed
through the
sharing of
information
across devices.
Zabbix Icinga OpenNMS WhatsUp Gold
SQUID is the core
memory usage
measurement used to
monitor other
processes. Status and
operations are well
created.
Deals specifically with
the critical services
loaded in the Linode,
external protocols are
easily monitored.
Traps the running
states of a service using
the SNMP within the
operation of any state
given.
Concerned with the
online web applications
and other windows
applications by
verifying the status of
each and checking its
health.
2.4 Bandwidth Monitoring
Nagios Core Ganglia CollectD Spiceworks Cacti
Uses SNMP for
complete
monitoring of
Custom metrics are
embedded here for
bandwidth
Uses the interface to
collect information
concerning all the
It comes along
with a
bandwidth
Uses
bandwidth in a
graph mode

Surname 6
the bandwidth
for both the
routers and
switches.
monitoring. Both the
local and remote
bandwidth can be
easily measured and
monitored.
traffic being
processed. All the
packets are described
in every plugin.
analyzer feature
that provides all
the
performance
historically.
where SNMP
polling allows
the box for
storing the
devices.
Zabbix Icinga OpenNMS WhatsUp Gold
Provides robust
monitoring for each
bandwidth used during
the operation.
With Linode, critical
services can be easily
monitored. Both the
internal host and the
external protocols can
be monitored with
ease.
Creates alert with
every operation
exceeding 1GB
bandwidth. This will
allow the user to make
own suggestions on
what to do next.
With the number of
users, this validates all
the business
operations using the
Cisco NBAR.
2.5 Resource Measuring
Nagios Core Ganglia CollectD Spiceworks Cacti
Other than
network services,
it also monitors
resources such as
processor loading
and the usage of
the disk.
It can measure the
overall host
resource such as
the grid and the
cluster computing.
Stores the
information
periodically in
some ERD files. All
the performance
metrics are well
stored.
Issues that arise
in the servers
and the
switches can be
easily spotted
and fixed with
the Spiceworks.
Data stored in
MySQL can be
easily measured
using the RRD
along with all the
devices in use.
Data is well
gathered properly.

Surname 7
Zabbix Icinga OpenNMS WhatsUp Gold
Ensures metrics
including networks, the
virtual machines, and
the servers.
Allows for distributed
monitoring of the
resources throughout
the whole process.
Spanning resources
multiple times is the
paramount factor in
this management tool.
All the state changes
are well analyzed and
allowed for.
2.6 Overall Availability
Nagios Core Ganglia CollectD Spiceworks Cacti
It is freely
available for
all Linux
system users.
It is scalable and
distributed
specifically for
higher
performance for
the computing
systems.
Throughout the
running of the
network. A
working operating
system allows the
collection of
information in all
sections of the
operating system.
Being a simple
network tool
for monitoring,
its real-time
updates on the
status, it is
preferably
available in
steady modes.
Being a fronted
application and stores
present statistics of the
network and the servers,
it is easily available
throughout the use in the
system.
Zabbix Icinga OpenNMS WhatsUp Gold
This is an open source
tool that allows low
consumption of the
overall available space
Just like a Nagios, it
provides a free source
integrated to fit all the
needs of the IT users.
Best fit the
administrators for
fixing and monitoring
issues with latencies.
Visible in all the
performances and the
status per the network.
3 Evaluation

Surname 8
Having mentioned the difference in the monitoring tools, it’s a critical factor to consider the best tool for
maximum uptime performance [2]. To fully benefit from these management tools, its best practice to
stay online both for maximizing the profits and for the evasion of the downtime performance. Pink
monitors and the SNMP devices allow the user to monitor the overall performance in line with the
notification as a result of a system fall. From the above, Nagios and the Zabbix are considered best for
the operation.
3.1 Nagios
This is a tool providing most centralized perspective in network setup and allows once to view the
services running whether they are up or already down [3].Through this process, one can get assurance
that he or she gets the alerts in a 24-hour system. With the addons well stated, monitoring and the
customization of the architecture s done based on several options.
3.2 Zabbix
This guarantees for all the uptime statuses. A service can be view as being a warning or at times at a
critical state, or when it is completely down and no longer running. In this case, the performance state is
deep within to allow measurements such as the temperature, power supply, the status of the fan and
the available space in the disk. With this information, the network administrator will be able to prevent
any given device within the network from failing. Other than the network protection, it also allows one
to check the availability of the bandwidth if sufficient or low [4].
4 Solution proposal
With the rampant occurrences of the server vulnerabilities, there is a need to safeguard the network
infrastructure with the networking monitoring tools. Servers need to be well managed. As a protection
mechanism, we are going to consider the use of the VPN networks and the Service Auditing.
4.1 VPN and the Private Networking
This is the best practice to prevent external users from accessing the network. Local computers can as
well be connected to the remote servers via the use of the VPN. Only the network that is used by the
servers can be accessed hence ensuring a proper full private network secure in nature [5].

Surname 9
4.2 Service Auditing
This is another best practice from where the security of a network can be improved. Other than having
the system running, the remaining task is to analyze the systems to identify well the areas that are
prone to attacks and those that can lock down the whole components. Services can be audited within
the infrastructure using a checklist stating all the service installed [6]. Through this, the administrator
will be able to tell which other services are running in the system other than the stated service
monitoring tools. Those that are running on different servers can be identified, moved to the respective
interface and set a firewall for blocking such confusion in the system.
5 Failure Prediction Impact
With the increase in network operations, there is a need to fully have a preventive measure in place.
Having identified the fault well, network administration will have some benefits like stated below:
a) Reduced costs-Maintenance costs resulting from system failure can be easily evaded with a well-
managed system.
b) Takes into account the quality of the overall performance. With all the service set in place, the
quality is considered for the future growth of the infrastructure.
c) One is also able to predict the analysis using the analytics about the attackers.
d) The time factor is well managed for future network capacities. The time that could be used in
analyzing the traffic is well managed elsewhere for other technological benefits.
6 Conclusion
From the above discussion, we can conclude that to have the best performance in an organization, we
need to have the monitoring tools of the top level for quick alerts. In this report we've compared the

Surname 10
various monitoring tools, even though there exists a difference, having at least one will guarantee safety
whole through the operation.

Surname 11
References
[1 J. Hernantes, G. Gallardo and N. Serrano, "IT Infrastructure-Monitoring Tools" in IEEE Software, vol. 32,
no. 04, pp. 88-93, 2015.doi: 10.1109/MS.2015.96
[2] M. Gangone, M. Whelan, K. Janoyan and L. Minnetyan, "Development of performance assessment
tools for a highway bridge resulting from controlled progressive monitoring", Structure and
Infrastructure Engineering, vol. 10, no. 5, pp. 551-567, 2013. Available: 10.1080/15732479.2012.757329.
[3] V. Fernández, A. Pazos, J. Saborido and M. Seco, "Arduino and Nagios integration for
monitoring", Journal of Physics: Conference Series, vol. 513, no. 6, p. 062015, 2014. Available:
10.1088/1742-6596/513/6/062015.
[4]P ool Wear and Failure Monitoring on an Open-Architecture Machine Tool", Precision Engineering, vol.
13, no. 3, pp. 237-238, 2010. Available: 10.1016/0141-6359(91)90203-u.
[5] K. Kim and H. Yang, "VPN (Virtual Private Network) SW's examination example analysis", Journal of
the Korea Academia-Industrial cooperation Society, vol. 11, no. 8, pp. 3012-3020, 2010. Available:
10.5762/kais.2010.11.8.3012.
[6] B. Morley, "Management Influence through a Quality Audit Service", Managerial Auditing Journal,
vol. 2, no. 3, pp. 17-20, 1987. Available: 10.1108/eb017599.