Benchmarking of Control Kernels on Open-Source RISC-V Processors

dc.contributorAalto-yliopistofi
dc.contributorAalto Universityen
dc.contributor.advisorPuri, Subash
dc.contributor.authorGhimire, Narayan
dc.contributor.schoolSähkötekniikan korkeakoulufi
dc.contributor.supervisorRyynänen, Jussi
dc.date.accessioned2023-01-29T18:04:56Z
dc.date.available2023-01-29T18:04:56Z
dc.date.issued2023-01-23
dc.description.abstractIn recent years, the RISC-V Instruction Set Architecture has emerged as an open-source alternative in the processor market which is dominated by proprietary architectures. The modern telecommunication industry has adopted the RISC-V architecture for accelerating communication data paths. In a telecommunication system, control kernels play a crucial role in managing the underlying hardware to direct the flow of information between devices. A control kernel typically configures the underlying infrastructure for the system and provides services such as scheduling, resource management, and data processing. Such tasks are heavily dependent on configurations of the 5G systems. This thesis presents a study of the telecommunication related control kernel’s performance and power efficiency on open-source RISC-V processors. The open-source RISC-V implementation, CV32E40P, maintained by OpenHW Group is selected for benchmarking against Nokia’s in-house processor core, NRISCV. The processors are synthesized at 1 GHz frequency for 7nm TSMC technology, and power is estimated on the synthesized cores using PowerArtist. The study finds that control kernels' performance and power consumption are largely influenced by the underlying microarchitecture of the RISC-V processor, with some control kernels achieving significantly better performance and power efficiency for specific implementations. This study provides insight into the strengths and weaknesses of different RISC-V processors for control kernel applications and can guide the design and implementation of future telecommunication systems.en
dc.format.extent56
dc.format.mimetypeapplication/pdfen
dc.identifier.urihttps://aaltodoc.aalto.fi/handle/123456789/119367
dc.identifier.urnURN:NBN:fi:aalto-202301291717
dc.language.isoenen
dc.locationP1fi
dc.programmeMaster’s Programme in Electronics and Nanotechnology (TS2013)fi
dc.programme.majorMicro- and Nanoelectronic Circuit Designfi
dc.programme.mcodeELEC3036fi
dc.subject.keywordRISC-Ven
dc.subject.keywordCV32E40Pen
dc.subject.keywordPPA analysisen
dc.subject.keywordcontrol kernelsen
dc.subject.keywordbenchmarksen
dc.titleBenchmarking of Control Kernels on Open-Source RISC-V Processorsen
dc.typeG2 Pro gradu, diplomityöfi
dc.type.ontasotMaster's thesisen
dc.type.ontasotDiplomityöfi
local.aalto.electroniconlyyes
local.aalto.openaccessyes
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
master_Ghimire_Narayan_2023.pdf
Size:
2.29 MB
Format:
Adobe Portable Document Format