Development of self-adaptaning virtual machines to improve learning outcomes in heterogeneous computing enviroment

[P. R. Stolić]

185 listova

Savremeno izvođenje nastavnih procesa u okviru visokoškolskog obrazovanja oslanja se na upotrebu principa učenja od kuće koji, pored prednosti, ispoljavaju i neke nedostatke. Jedan od uočenih nedostataka jeste heterogenost računarskog okruženja, kao rezultat upotrebe ličnih računarskih resursa studenata budući da ti računarski resursi međusobno ispoljavaju veliku raznolikost. Način prevazilaženja nastale heterogenosti jeste prevođenje u homogeni oblik upotrebom virtuelnih mašina. Ovakvim pristupom vrši se približavanje postizanju uniformnosti nastavnog procesa posmatrano sa stanovišta upotrebe računarske tehnike. Virtuelizacija unosi novu kompleksnost jer se sve aktivnosti virtuelizacije sprovode od strane studenata i nastavnog osoblja. To podrazumeva da će se akteri nastavnog procesa privremeno udaljiti od primarnih ciljeva vezanih za postizanje pozitivnih ishoda učenja. Potrebno je razviti mehanizme koji će omogućiti automatizaciju procesa vezanih za primenu virtuelne mašine u nastavnom procesu u cilju svođenja učešća aktera nastavnog procesa u virtuelizaciji na prihvatljiv minimum. Ovo je postignuto razvojem samoadaptirajuće virtuelne mašine koja, u odnosu na klasičnu virtuelnu mašinu, poseduje adaptivna svojstva u hardverskom i softverskom smislu. Adaptivnost u hardverskom smislu ogleda se u automatskom prilagođavanju trenutnim uslovima okruženja u kome virtuelna mašina ostvaruje rad, analizom parametara okruženja, statističkom obradom istih i predikcijom budućih stanja, čime se definišu parametri rada virtuelne mašine. U softverskom smislu, adaptivnost se postiže korišćenjem principa ontološkog inženjerstva čime se svi instalacioni procesi u okviru virtuelne mašine obavljaju bez direktnog učešća korisnika. Sve funkcionalnosti realizovanog rešenja potvrđene su odgovarajućim testiranjima.

The modern implementation of teaching processes within higher education relies on the use of the principles of learning from home, which have some disadvantages. One of the observed disadvantages is the heterogeneity of the computer environment, as a result of the use of personal computer resources of students, since these computer resources exhibit great diversity among themselves. The way to overcome the resulting heterogeneity is to translate it into a homogeneous form using virtual machines. Virtualization introduces new complexity because all virtualization activities are performed by students and teaching personnel. This means that the actors of the teaching process will temporarily move away from the primary goals related to achieving positive learning outcomes. It is necessary to develop mechanisms that will enable the automation of processes related to the application of a virtual machine in the teaching process in order to reduce the participation of actors of the teaching process in virtualization to an acceptable minimum. This was achieved by developing a self-adaptive virtual machine that, compared to a classic virtual machine, has hardware and software adaptive properties. Hardware adaptability is reflected in automatic adaptation to the current conditions of the environment in which the virtual machine works, by analyzing the parameters of the environment, statistical processing of them and prediction of future states, which defines the operating parameters of the virtual machine. Software adaptability is achieved by using the principles of ontological engineering, whereby all installation processes within the virtual machine are performed without direct user participation. All functionalities of the implemented solution were confirmed by appropriate tests.

adaptivnost, virtuelizacija, virtuelna mašina, virtuelni hardver, visokoškolsko obrazovanje, ontološko inženjerstvo, prediktivna analitika, softver, hardver

adaptivity, hardware, higher education, ontology engineering, predictive analytics, software, virtualization, virtual machine, virtual hardware

378.018.43:004.4/.5(043.3)

Tekst