Klessydra-T: Designing Vector Coprocessors for Multithreaded Edge-Computing Cores
Abdallah Cheikh, Stefano Sordillo, Antonio Mastrandrea, Francesco Menichelli, Giuseppe Scotti, Mauro Olivieri
2021IEEE Micro38 citationsDOIOpen Access PDF Abstract
Computation-intensive kernels, such as convolutions, matrix multiplication, and Fourier transform, are fundamental to edge-computing AI, signal processing, and cryptographic applications. Interleaved-multithreading (IMT) processor cores are interesting to pursue energy efficiency and low hardware cost for edge computing, yet they need hardware acceleration schemes to run heavy computational workloads. Following a vector approach to accelerate computations, this article explores possible alternatives to implement vector coprocessing units in RISC-V cores, showing the synergy between IMT and data-level parallelism in the target workloads.
Topics & Concepts
Computer scienceCoprocessorParallelism (grammar)AccelerationParallel computingHardware accelerationMulti-core processorEfficient energy useSignal processingCryptographyEnhanced Data Rates for GSM EvolutionMatrix (chemical analysis)Vector processorEmbedded systemComputer hardwareFast Fourier transformEnergy (signal processing)Parallel processingComputational scienceKernel (algebra)Instruction setDigital signal processingSide channel attackField-programmable gate arrayEncoding (memory)Matrix multiplicationMultithreadingChipSystem on a chipAlgorithm designReduction (mathematics)Computer architectureFourier transformEnergy consumptionTask (project management)ThroughputDigital signal processorTask parallelismParallel Computing and Optimization TechniquesNumerical Methods and AlgorithmsCloud Computing and Resource Management