Performance Evaluation and Analysis for Block I/O Access Pattern between KVM-based Virtual Machine and Real Machine in the Virtualized Environment

The main obstacle in mass adoption of cloud computing for database operations is the data security issue. In this paper, it is shown that IT services particularly in hardware performance evaluation invirtual machinecan be accomplished effectively without IT personnel gaining access to real data for diagnostic and remediation purposes. The proposed mechanisms utilizedTPC-Hbenchmark to achieve 2 objectives. First, the underlying hardware performance and consistency is supervised via a control system, which is constructed using a combination ofTPC-Hqueries,linear regression, andmachine learningtechniques. Second,linear programmingtechniques are employed to provide input to the algorithms that construct stress-testing scenarios in thevirtual machine, using the combination ofTPC-Hqueries. These stress-testing scenarios serve 2 purposes. They provide the boundary resource threshold verification to the first control system, so that periodic training of the synthetic data sets for performance evaluation is not constrained by hardware inadequacy, particularly when the resources in thevirtual machineare scaled up or down which results in the change of the utilization threshold. Secondly, they provide a platform for response time verification on critical transactions, so that the expected Quality of Service (QoS) from these transactions is assured.

Download Full-text

PENGAMANAN NESTED VIRTUALIZATION SERVER UNTUK E-LEARNING MOODLE PADA PROXMOX VE MENGGUNAKAN METODE HIGH AVAILABILITY

BUANA ILMU ◽

10.36805/bi.v6i1.1982 ◽

2021 ◽

Vol 6 (1) ◽

pp. 20-28

Author(s):

Asep Sofyan Wahyudin ◽

Muhamad Abdul Aziz ◽

Eryan Ahmad Firdaus ◽

Kanggep Andrijana Kusuma

Keyword(s):

Information System ◽

Virtual Machine ◽

Virtual Machines ◽

High Availability ◽

Computing Technology ◽

E Learning ◽

Optimal Information ◽

Real Machine

Perkembangan teknologi komputasi saat ini luar biasa sangan pesat, tak sedikit sebuah server yang dulunya hanya dimanfaatkan untuk menangani satu layanan saja, kini hal itu sudah di anggap sebagai pemborosan resource, salah satu penyebabnya adalah perkembangan metode virtualisasi yang di anggap dapat mengoptimalkan sebuah server untuk menangani lebih banyak layanan dengan cara membuat beberapa mesin virtual dalam satu mesin real untuk memaksimalkan kemampuan server, lalu apakan dalam sebuah mesin virtual dapat menjalakan mesin virtual lagi, amankah bila digunakan untuk menangani layanan sistem informasi secara optimal. Kata Kunci- Virtualisasi Bersarang, E-Learning, High Availability. The development of computing technology is currently very fast, not a few servers that were previously only used to handle one service, now it is considered a waste of resources, one of the causes is the development of virtualization methods which are considered to optimize servers for more services by creating several virtual machines in to one real machine to maximize server capabilities, so in a virtual machine can it run the virtual machine again, is it safe to using it for optimal information system services. Keyword- Nested Virtualization, E-Learning, High Availability.

Download Full-text

A Cloud-based Middleware for Self-Adaptive IoT-Collaboration Services

Sensors ◽

10.3390/s19204559 ◽

2019 ◽

Vol 19 (20) ◽

pp. 4559

Author(s):

Park ◽

Park

Keyword(s):

Performance Evaluation ◽

Virtual Machine ◽

Context Awareness ◽

Efficient Solutions ◽

Cloud Services ◽

Single System ◽

Considerable Time ◽

Web Based ◽

Criminal Suspect ◽

Self Adaptive

The middleware framework for IoT collaboration services should provide efficient solutions to context awareness and uncertainty issues among multiple collaboration domains. However, existing middleware frameworks are mostly limited to a single system, and developing self-adaptive IoT collaboration services using existing frameworks requires developers to take considerable time and effort. Furthermore, the developed IoT collaboration services are often dependent on a particular domain, which cannot easily be referenced in other domains. This paper proposes a cloud-based middleware framework that provides a set of cloud services for self-adaptive IoT collaboration services. The proposed middleware framework is generic in the sense that it clearly separates domain-dependent components from the layers that leverage existing middleware frameworks. In addition, the proposed framework allows developers to upload domain-dependent components onto the cloud, search for registered components, and launch Virtual Machine (VM) running a new MAPE cycle via a convenient web-based interface. The feasibility of the proposed framework has been shown with a simulation of an IoT collaboration service that traces a criminal suspect. The performance evaluation shows that the proposed middleware framework runs with an overhead of only 6% compared to pure Java-based middleware and is scalable as the number of VMs increases up to 16.

Download Full-text