Rodinia: A benchmark suite for heterogeneous computing

By Shuai Che, Michael Boyer, Jiayuan Meng, David Tarjan, Jeremy W. Sheaffer, Sang-Ha Lee, Kevin Skadron. 2009

3,102 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

Property Inference Attacks on Fully Connected Neural Networks using Permutation Invariant Representations (2018, methodology): “We used perf with a sampling rate of 2 seconds to profile 15 cryptocurrency mining applications and 19 non-mining applications from the Rodinia [9] and Parboil [40] benchmark suites.”
DORA: Optimizing Smartphone Energy Efficiency and Web Browser Performance under Interference (2018, background): “The algorithms behind these workloads form the basic building blocks of current and future smartphone workloads, representing sensor data analysis, image processing, thermal prediction and management, video games, and medical applications [44].”
MASK: Redesigning the GPU Memory Hierarchy to Support Multi-Application Concurrency (2018, methodology): “We randomly select 27 applications from the CUDA SDK [94], Rodinia [31], Parboil [121], LULESH [65, 66], and SHOC [37] suites.”
MachSuite: Benchmarks for accelerator design and customized architectures (2014, background): “Many high-quality benchmarks have emerged from the GPGPU community in the last several years [9], [18], [44].”
End-to-End Deep Learning of Optimization Heuristics (2017, neutral): “3 7 114 Rodinia [31] 3.”
Scalable GPU graph traversal (2012, background): “It is also representative of a class of parallel computations whose memory accesses and work distribution are both irregular and data-dependent.”
Parboil: A Revised Benchmark Suite for Scientific and Commercial Throughput Computing (2012, methodology): “The Rodinia benchmarks published by the University of Virgi nia [2] are very similar in philosophy and development to the Parboil benchmarks.”
Benchmarking TPU, GPU, and CPU Platforms for Deep Learning (2019, background): “Others compare cloud computing providers [34], heterogeneous platforms [10], and cloud support for HPC [22].”
Cache-Conscious Wavefront Scheduling (2012, methodology): “BFS Graph Traversal [9] BFS Kmeans [9] KMN Memcached [16] MEMC Garbage Collection [5, 36] GC Moderately Cache Sensitive (MCS) Name Abbr.”
Zorua: A holistic approach to resource virtualization in GPUs (2016, methodology): “resulting utilization of physical resources on GPU performance, we conduct an experiment on a Maxwell GPU system (GTX 745) with 20 GPGPU workloads from the CUDA SDK [52], Rodinia [9], GPGPU-Sim benchmarks [5], Lonestar [8], Parboil [66], and US DoE application suites [72].”
A characterization of the Rodinia benchmark suite with comparison to contemporary CMP workloads (2010, methodology): “How difficult is an application to implement [8]–[10]?”
Neural Code Comprehension: A Learnable Representation of Code Semantics (2018, neutral): “Lines Size Pairs Machine Learning Tensorflow [1] 2,492 16,943,893 220,554 260,250,973 High-Performance Computing AMD APP SDK [9] 123 1,304,669 4,146 45,081,359 BLAS [22] 300 280,782 566 283,856 Benchmarks NAS [57] 268 572,521 1,793 1,701,968 Parboil [59] 151 118,575 2,175 151,916 PolybenchGPU [27] 4”
Mining on Someone Else's Dime: Mitigating Covert Mining Operations in Clouds and Enterprises (2017, methodology): “Figure 2 presents this comparison between mining software and some popular and common GPU workloads taken from the Rodinia [33] and Parboil [49] GPU-benchmark suites.”
TOP-PIM: throughput-oriented programmable processing in memory (2014, methodology): “Rodinia is a widely used GPGPU benchmark suite intended to be a representative sample covering a wide range of workload characteristics [18].”
A Comprehensive Performance Comparison of CUDA and OpenCL (2011, methodology): “Benchmarks are selected from the SHOC benchmark suite, NVIDIA's SDK, and the Rodinia benchmark suite [22].”
NDA: Near-DRAM acceleration architecture leveraging commodity DRAM devices and standard memory modules (2015, background): “SRAD Speckle reducing anisotropic diffusion [52] 3 Yes 99.”
Accelerating CUDA graph algorithms at maximum warp (2011, methodology): “Backprop [12], a neural network application, performs vector-matrix multiplication in its first kernel.”
Analyzing Machine Learning Workloads Using a Detailed GPU Simulator (2018, methodology): “Although the existing GPGPU-Sim simulator correctly runs a large number of CUDA applications such as those found in Rodinia [38], machine learning workloads presented additional complexities.”
A Survey of CPU-GPU Heterogeneous Computing Techniques (2015, neutral): “The choice of programs in Rodinia [Che et al. 2009] is based on Berkeley’s dwarf taxonomy and is aimed at covering different types of parallel communication patterns and synchronization techniques.”