Optimal dynamic resource allocation for heterogenous cloud data centers

Download
2019
Ekici, Nazım Umut
Today's data centers are mostly cloud-based with virtualized servers to provide on-demand scalability and flexibility of the available resources such as CPU, memory, data storage and network bandwidth. Heterogeneous cloud data centers (CDCs) offer hardware accelerators in addition to these standard cloud server resources. A cloud data center provider may provide Infrastructure as a Service and Platform as a Service (IPaaS), where the user gets a virtual machine (VM) with processing, memory, storage and networking resources, which can be installed with any desired operating system and software. Differently, Software as a Service (SaaS), only enables user access to provided application for example via a web browser without any control of the underlying infrastructure. In this context, it is important to note that the data processing for SaaS can be executed on different physical resources such as a server as well as a hardware accelerator with different performance and power consumption. To this end, a very significant feature of heterogeneous CDCs is that they offer the flexibility of meeting user demands for SaaS by choosing among the available physical resource alternatives. To utilize this flexibility, a CDC resource manager must decide which resource alternative will be chosen, along with the decision of the physical resource the request will be assigned to. In this thesis we propose ACCLOUD-MAN (ACCelerated CLOUD MANager), a novel resource manager for heterogeneous CDCs. ACCLOUD-MAN’s resource management objective is to reduce the power consumption of the CDC in order to support green computing. To this end, the resource allocation problem is modeled as an integer linear programming problem and is implemented in MATLAB, along with a cloud data center simulation platform. We evaluate the performance of ACCLOUD-MAN under different realistic cloud workloads. Simulation results show that the proposed ACCLOUD-MAN outperforms existing resource allocation methods such as OpenStack.

Suggestions

Generalized resource management for heterogeneous cloud data centers
Erol, Ahmet; Güran Schmidt, Şenan Ece.; Department of Electrical and Electronics Engineering (2019)
OpenStack is a widely used management tool for cloud computing which is designed to work on servers and allocate standard computing resources such as CPU, memory or disk. The current trend for integrating different hardware accelerators such as FPGAs and GPUs in the cloud requires managing these heterogeneous resources. In this thesis, we propose a generalization for OpenStack Nova project which extends the relevant data structures to include these new resources. More importantly, we present a new lightweig...
EXTENSION OF AN OPEN SOURCE RESOURCE MANAGEMENT TOOL FOR HETEROGENEOUS CLOUD DATA CENTERS: IMPLEMENTATION AND EVALUATION
Doğan, Taha; Schmidt, Şenan Ece; Department of Electrical and Electronics Engineering (2022-2-11)
Cloud Computing is enabled by the virtualization of computing resources to realize users' requests of virtual machines (VMs) and data processing in the scope of Infrastructure as a Service (IaaS) and Software as a Service (SaaS) respectively. The current heterogeneous cloud data centers incorporate hardware accelerators in addition to the conventional servers to offer these services more efficiently. It is an important research problem to allocate heterogeneous physical computing resources to a mixture of ...
Improving Hadoop Hive Query Response Times Through Efficient Virtual Resource Allocation
Dokeroglu, Tansel; Cinar, Muhammet Serkan; SERT, SEYYİT ALPER; Coşar, Ahmet; Yazıcı, Adnan (2015-10-28)
The performance of the MapReduce-based Cloud data warehouses mainly depends on the virtual hardware resources allocated. Most of the time, the resources are values selected/given by the Cloud service providers. However, setting the right virtual resources in accordance with the workload demands of a query, such as the number of CPUs, the size of RAM, and the network bandwidth, will improve the response time when querying large data on an optimized system. In this study, we carried out a set of experiments w...
Optimization of XNOR Convolution for Binary Convolutional Neural Networks on GPU
Kaya, Mete Can; İnci, Alperen; Temizel, Alptekin (2020-10-09)
Binary convolutional networks have lower computational load and lower memory foot-print compared to their full-precision counterparts. So, they are a feasible alternative for the deployment of computer vision applications on limited capacity embedded devices. Once trained on less resourceconstrained computational environments, they can be deployed for real-time inference on such devices. In this study, we propose an implementation of binary convolutional network inference on GPU by focusing on optimization ...
A Quality model for cloud-based enterprise information systems
Şener, Umut; Eren, Pekin Erhan; Department of Information Systems (2016)
Organizations have migrated from on-premise enterprise information systems to Cloud-based Enterprise Information Systems (Cloud-EIS) due to the benefits of cloud computing, such as flexibility, availability on demand, and interdependence in information technology infrastructure. Accordingly, enterprises perceive the significance of the quality of Cloud-EIS for improving their businesses, and they pay more attention to selecting the suitable Cloud-EIS. Having looked at the extensive literature, only a few re...
Citation Formats
N. U. Ekici, “Optimal dynamic resource allocation for heterogenous cloud data centers,” Thesis (M.S.) -- Graduate School of Natural and Applied Sciences. Electrical and Electronics Engineering., Middle East Technical University, 2019.