Identifikasi terjadinya penyimpangan pada lingkungan virtual dengan metode adaptive windowing

Nathaniel, Farrell (2021) Identifikasi terjadinya penyimpangan pada lingkungan virtual dengan metode adaptive windowing. Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title (70).pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (847kB)
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (302kB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (936kB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (753kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (325kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (391kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (176kB)

Abstract

Concept Drift is a phenomena in data science which occurs when data changes in an unexpected way due to user or operational activity. This occurence has to be observed at all times, especially on data stream, case example, a scalability process in cloud computing where scaling happens in real-time, data needs to be identified whether a drift happens when a scaling activity is done. Any relation between VM scaling process and drifting data will be observed. Hardware virtualization process could have caused the system to output inaccurate data. The study will use CloudSim Plus, a fork of CloudSim 3. Using the provided simulation Vertical VM Scaling, the data will be outputted to a .csv file and the text will cleansed. Once only the CPU Usage values remain, the data will be tested on drift detection algorithm ADWIN, provided by Python module scikit-multiflow. After being tested on ADWIN, drift is detected on simulation data. Analysis found that after VM switches Cloudlet, system will start outputting 0% values. This is caused by system having to wait for the simulation to finish switching the Cloudlet, and the system end up having to wait until the next iteration to output the next data./Concept Drift adalah fenomena dalam data science yang terjadi ketika data berubah seiring waktu karena suatu aktivitas pengguna atau operasional. Hal ini perlu diamati pada data stream, pada contoh kasus proses skalabilitas dalam cloud computing di mana proses skalabilitas terjadi secara real-time, maka perlu diidentifikasi apakah terjadi penyimpangan pada data ketika aktivitas ini dilakukan. Juga akan diperhatikan jika adakah hubungan antara proses skalabilitas VM dan data yang menyimpang. Proses virtualisasi hardware dapat menyebabkan sistem tidak mengeluarkan data yang akurat. Penelitian ini akan menggunakan CloudSim Plus, fork dari CloudSim. Menggunakan contoh simulasi scaling VM secara vertikal, data akan dioutput ke .csv lalu dicleanse output textnya. Setelah hanya tersisa nilai penggunaan CPU dari scaling vertikal, data diuji dengan algoritma deteksi drift ADWIN, menggunakan module Python scikit-multiflow. Setelah diuji dengan ADWIN, drift terdeteksi dalam data simulasi. Dianalisa bahwa ketika terjadi pergantian Cloudlet dalam VM, sistem tidak mengeluarkan data yang benar, tetapi banyak mengeluarkan data bernilai 0% yang terjadi setelah sistem melakukan pergantian Cloudlet. Ini disebabkan oleh sistem yang terlambat mencatat nilai penggunaan CPU, karena simulasi harus menunggu pergantian Cloudletnya selesai sebelum memunculkan data selanjutnya.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Nathaniel, FarrellNIM01082170002farrellnathaniel25@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorMurwantara, I MadeNIDN0302057305made.murwantara@uph.edu
Thesis advisorPanduwinata, FransNIDN0306028201frans.panduwinata@uph.edu
Uncontrolled Keywords: cloud computing; concept drift; adwin; virtualisasi; vm; simulasi
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics
Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics
Depositing User: Users 9551 not found.
Date Deposited: 01 Mar 2021 08:33
Last Modified: 01 Mar 2021 08:33
URI: http://repository.uph.edu/id/eprint/24976

Actions (login required)

View Item View Item