A Workflow for Offering Hardware Accelerators as a Cloud Computing Service: Implementation and Evaluation

Download

10450923.pdf

Date

2022-2

Author

Tırlıoğlu, Anıl

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

310
views

503
downloads

Cloud computing and hardware accelerators are two paradigm changes in the field of information technologies and computers. Accordingly, this thesis proposes a workflow for offering users hardware accelerators implemented on FPGA as computing resources in a heterogeneous cloud data center. To this end, we perform the virtualization of FPGA resources as reconfigurable regions (RRs) and provide these resources through OpenStack, an open-source cloud resource management platform. Our workflow is designed for SoC FPGA platforms with a processor. The OpenStack module in the SoC processor is implemented as embedded software that works with other OpenStack modules. An accelerator image selected by the user can be programmed to an RR through OpenStack. The FPGA platform in our architecture features 40 Gbps Ethernet IP Cores and an on-chip switch that enables the communication of the RRs with each other, the SoC processor and the 40 Gbps Ethernet. To this end, distributed accelerator implementations can be realized on the same FPGA, and data received from the users can be processed and delivered back to the user. We explore OpenCL-based accelerator realization, which defines the data exchange between the accelerators and the CPU software. Furthermore, we demonstrate the implementation of the accelerators as stand-alone IP cores together with their wrappers in a more custom design flow. We implement the entire workflow on the Xilinx ZC706 board. The functional correctness and performance experiments are conducted throughout the thesis work. The experiments cover the effects of the Ethernet interface on the performance in accordance with the cloud data center operation.

Subject Keywords

Hardware accelerator, hardware accelerated cloud data center, partial reconfiguration, FPGA virtualization

URI

https://hdl.handle.net/11511/96380

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

A Software Development Process Model for Cloud by Combining Traditional Approaches Hacaloglu, Tuna; Eren, Pekin Erhan; Mishra, Deepti; Mishra, Alok (2015-10-30) Even though cloud computing is a technological paradigm that has been adopted more and more in various domains, there are few studies investigating the software development lifecycle in cloud computing applications and there is still not a comprehensive software development process model developed for cloud computing yet. Due to the nature of cloud computing that is completely different from the traditional software development, there is a need of suggesting process models to perform the software developmen...
Hardware Accelerators for Cloud Computing: Features and Implementation Tirlioglu, Anil; Demir, Omer Bayram; Yazar, Alper; Schmidt, Şenan Ece (2021-01-01) In this paper, hardware accelerator (FHA) applications realized on FPGA that can be offered as a service in cloud computing systems are discussed. It is necessary to know the hardware resources used by FHA applications and the performance they provide for the efficient meeting of the user requests and effective resource planning. To this end, the first contribution of this paper is to provide a compilation of the literature on the features of frequently used hardware accelerators (matrix multiplication, fac...
A Big Data Analytical Approach to Cloud Intrusion Detection Gulmez, Halim Gorkem; Tuncel, Emrah; Angın, Pelin (null; 2018-06-30) Advances in cloud computing in the past decade have made it a feasible option for the high performance computing and mass storage needs of many enterprises due to the low startup and management costs. Due to this prevalent use, cloud systems have become hot targets for attackers aiming to disrupt reliable operation of large enterprise systems. The variety of attacks launched on cloud systems, including zero-day attacks that these systems are not prepared for, call for a unified approach for real-time detect...
A Distributed Monitoring and Reconfiguration Approach for Adaptive Network Computing Bhargava, Bharat; Angın, Pelin; Ranchal, Rohit; Lingayat, Sunil (2015-01-01) The past decade has witnessed immense developments in the field of network computing thanks to the rise of the cloud computing paradigm, which enables shared access to a wealth of computing and storage resources without needing to own them. While cloud computing facilitates on-demand deployment, mobility and collaboration of services, mechanisms for enforcing security and performance constraints when accessing cloud services are still at an immature state. The highly dynamic nature of networks and clouds ma...
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

A. Tırlıoğlu, “A Workflow for Offering Hardware Accelerators as a Cloud Computing Service: Implementation and Evaluation,” M.S. - Master of Science, Middle East Technical University, 2022.