A NOVEL FLEXIBLE ON-CHIP SWITCH ARCHITECTURE FOR RECONFIGURABLE HARDWARE ACCELERATORS

Download

Thesis_FatihYazici_13082021.pdf

Date

2021-8-13

Author

Yazıcı, Fatih

Metadata

Show full item record

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

Item Usage Stats

321
views

106
downloads

This thesis work proposes ReFlex Switch, a novel, scalable on-chip packet switch architecture, that is designed to interconnect heterogeneous IP cores at high speeds. One target application for ReFlex switch is hardware accelerated cloud computing where the cloud servers feature FPGA cards with reconfigurable regions to implement accelerators demanded by the users. In this setting, the increasing data rates call for line-speed operation of the on-chip switch to maintain scalability. The first requirement of the line rate operation is a fabric arbiter design to achieve maximal throughput while allocating the switching bandwidth as required among contending traffic sources. The second requirement together with the limited on-chip memory resources is the efficient buffer management at the switch inputs. Furthermore, the service needs of the applications change with the realized accelerators on the FPGA card. ReFlex Switch fulfills these requirements with an on-chip hardware architecture that is flexibly configured according to the specified system parameters. To this end, this thesis proposes Credit ARbiter (CAR), a novel fabric arbiter with Quality of Service (QoS) support and ReFlex Buffer Management (ReFBM) a novel input buffer organization for buffer allocation to the connected cores according to their traffic demand. ReFlex Switch is implemented on Xilinx XC7Z100 SoC FPGA at 40 Gbps line rate, with CAR and ReFBM together with legacy arbiters and buffer organizations for resource use and performance comparison. The evaluations demonstrate that ReFlex switch can be flexibly instantiated with different memory parameters, arbiters and memory organizations. Furthermore, CAR and ReFBM achieve desired performance goals and outperform comparable work in the literature.

Subject Keywords

on-chip switch, switch fabric arbitration, dynamic buffer management, hardware accelerated cloud data center, partial reconfiguration

URI

https://hdl.handle.net/11511/91660

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

An On-chip Switch Architecture for Hardware Accelerated Cloud Computing Systems Donanim Hizlandiricili Bulut Bilisim Sistemleri icin Yonga-ustu Anahtar Mimarisi Yazıcı, Fatih; Yildiz, Ayhan Sefa; Yazar, Alper; Schmidt, Şenan Ece (2020-10-05) © 2020 IEEE.In this paper, we propose a scalable on-chip packet switch architecture for hardware accelerated cloud computing systems. Our proposed switch architecture is implemented on the FPGA and interconnects reconfigurable regions, 40 Gbps Ethernet interfaces and a PCIe interface. The switch fabric operates at line speed to achieve scalability. We propose a new algorithm that grants access to the fabric according to the allocated prioritization to input-output port pairs. The switch is implemented on Xi...
DESIGN, IMPLEMENTATION AND VERIFICATION OF A HIGH-SPEED ON-CHIP PACKET SWITCH YILDIZ, Ayhan Sefa; Schmidt, Şenan Ece; Department of Electrical and Electronics Engineering (2022-2-9) In this thesis, an on-chip packet switch architecture to interconnect modules on System on Chip (SoC) platforms at high line speeds is proposed. The particular target application for the proposed on-chip switch is hardware accelerated cloud computing systems. To this end, FPGA Accelerator Cards (FAC) are employed in heterogeneous cloud data centers which implement hardware accelerators on the FPGA. The data from the cloud user is brought on the accelerators and delivered after processing through high-speed ...
A Digitally programmable application specific integrated circuit for drive and data acquisition of imaging sensorsMethod of moments analysis of slotted waveguide antenna arrays Bayhan, Nusret; Akın, Tayfun; Eminoğlu, Selim; Department of Electrical and Electronics Engineering (2014) This thesis explains the implementation of a digital programmable Application Specific Integrated Circuit (ASIC) designed for imaging applications. The primary function of this ASIC is to drive imaging sensors and to do basic processing on the digital video data coming from the sensors. The ASIC is designed to handle the communication between the imaging sensor and the system. Using command based high-level instructions, this two-way communication is simplified. The ASIC can also be used to store and update...
A New Hardware-in-the-Loop Simulator for CNC Machine Applications USENMEZ, S.; MUTLU, B. R.; Yaman, Ulaş; KILIÇ, ERGİN; Dölen, Melik; Koku, Ahmet Buğra (2013-03-01) This study focuses on an integrated software and hardware platform that is capable of performing (real-time/nonreal- time) hardware-in-the-loop simulation of dynamic systems, including electrical machinery, CNC machine tools. In this approach, once the dynamics of the plant to be controlled is defined via C++ language, the resulting code is cross-compiled automatically on a PC. Executable files along with the necessary drivers are downloaded onto the composite hardware platform that consists of a Field Prog...
A Distributed Fault-Tolerant Topology Control Algorithm for Heterogeneous Wireless Sensor Networks Bagci, Hakki; KÖRPEOĞLU, İBRAHİM; Yazıcı, Adnan (Institute of Electrical and Electronics Engineers (IEEE), 2015-04-01) This paper introduces a distributed fault-tolerant topology control algorithm, called the Disjoint Path Vector (DPV), for heterogeneous wireless sensor networks composed of a large number of sensor nodes with limited energy and computing capability and several supernodes with unlimited energy resources. The DPV algorithm addresses the k-degree Anycast Topology Control problem where the main objective is to assign each sensor's transmission range such that each has at least k-vertex-disjoint paths to superno...

Citation Formats

F. Yazıcı, “A NOVEL FLEXIBLE ON-CHIP SWITCH ARCHITECTURE FOR RECONFIGURABLE HARDWARE ACCELERATORS,” M.S. - Master of Science, Middle East Technical University, 2021.