60 % which may result in overheating of the system, decrease frame rates and even cause system
crashes.
Thus, the player must be made aware if the game has system requirements not compatible with his
system. A good RAM specification is needed for a smooth experience while playing games. This
ensures that there are fewer lags and allows for higher frame rates. RAM usage also depends on the
kind of software or game that is being run on the user’s device. Hence, this paper also proposes an
additional feature that will help determine if the user can run a particular game on his or her device
based on his system requirements.
The paper has the following structure, the ‘Literature Survey’ Section provides a concise overview of
the existing literature and tools available for the analysis of CPU metrics, the next section –
‘Comparison Table’ illustrates the different tables and diagrams used to support the ideas presented in
this paper, ‘Proposed Work’ section discusses the methodology of implementation in detail while the
results and outputs are presented in the ‘Results and Discussions’ section. Finally the paper discusses
some aspects of improvement in the future and provides a conclusion outlining the key points of the
entire paper.
2.1. LITERATURE SURVEY
Stefanov and Gradskov (2016) monitored and analysed some properties of CPU usage data provided
by Linux Kernel. This work analysed CPU usage data provided by the Linux kernel and how CPU
load level is calculated based on these data. For every active CPU in the system, the kernel gives the
amount of time, measured in 1/100th of a second, that the system spent in different modes of
execution since boot. These different modes are ‘user mode’ (running user processes), ‘user mode with
low priority’ (nice), ‘system mode’ (running kernel), ‘idle’, ‘iowait’ (idle while waiting for IO request
to complete), ‘irq’ (processing interrupts), ‘softirq’ (processing software interrupts).
Formula for calculating Load Level (L): If ‘Tim’ is the time spent in ‘m-th’ mode at time moment ‘i’
then, the CPU load level ‘Lm’ for the mode m is represented below using Equation (1):
(1)
Urriza and Clariño (2021) used Python to web scrape reviews written on ‘Steam’ website and
classifies them into either Audio, Gameplay or Graphics. Further it also categorizes them into Positive,
Negative and Neutral sentiments.
Gomes and Correia (2020) Cryptojacking infects the browsers and does CPU intensive computation to
mine cryptocurrencies on behalf of cyber criminals. The paper introduces a new Cryptojacking
detection mechanism based on monitoring CPU usage of visited web pages. They use machine
learning along with monitoring and monitoring the precision and recall values to make a decision. In
this Study they used CPU utilization for decking if certain games can be run or not on the computer,
similarly here we are utilizing the CPU for detecting cryptojacking. It uses ‘mpstat’, a command line
tool to monitor the CPU. It gives us several metrics such as CPU Utilization at every level, CPU
Utilization at user level, CPU Utilization at system level, time that the CPU or CPUs were idle during
which the system had an outstanding disk I/O request, time spent by the CPU or CPUs servicing
hardware interrupts, time spent by the CPU or CPUs servicing software interrupts, time spent in
involuntary wait by the virtual CPU or CPUs while the hypervisor was servicing another virtual
processor, time spent by the CPU or CPUs running a virtual processor; time that the CPU or CPUs
were idle and the system did not have an outstanding disk I/O request.
“Windows Task Manager” can optimize performance by terminating programmes and processes,
changing processing priorities, and setting processor affinity as necessary. Task Manager shows
m=
m
m
(Ti
- Ti-1
)
https://doi.org/10.17993/3ctic.2022.112.239-250
3C TIC. Cuadernos de desarrollo aplicados a las TIC. ISSN: 2254-6529
Ed. 41 Vol. 11 N.º 2 August - December 2022
242