The PARSEC benchmark suite: Characterization and architectural implications

By Christian Bienia, Sanjeev Kumar, Jaswinder Pal Singh, Kai Li. International Conference on Parallel Architectures and Compilation Techniques, 2008

3,813 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/588dda88f15f1622ee7de7631f2824c23eea60df

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

Rodinia: A benchmark suite for heterogeneous computing (2009, methodology): “The original code is from the Parsec Benchmark suite developed by Princeton University [2].”
Benchmarking modern multiprocessors (2011, background): “The work presented in this chapter was previously published in [10, 12].”
DianNao: a small-footprint high-throughput accelerator for ubiquitous machine-learning (2014, background): “This trend even starts to percolate in our community where it turns out that about half of the benchmarks of PARSEC [2], a suite partly introduced to highlight the emergence of new types of applications, can be implemented using machine-learning algorithms [4].”
DRAMSim2: A Cycle Accurate Memory System Simulator (2011, methodology): “Fig.4 shows the power consumption of a DRAMSim2 simulation of the PARSEC benchmark “fluidanimate”.”
STAMP: Stanford Transactional Applications for Multi-Processing (2008, background): “PARSEC [4] is a very recent parallel benchmark that was created to address some of the shortcomings in the abovementioned benchmarks.”
Sniper: Exploring the level of abstraction for scalable and accurate parallel multi-core simulation (2011, methodology): “To illustrate the case for heterogeneous workloads, we consider the dedup benchmark from the PARSEC benchmark suite [3].”
Scaling the bandwidth wall: challenges in and avenues for CMP scaling (2009, methodology): “To illustrate this case, we run PARSEC [5] benchmarks on a shared L2 cache multicore simulator, and measure the actual fraction of data sharing.”
Outstanding Research Problems in NoC Design: System, Microarchitecture, and Circuit Perspectives (2009, background): “Similarly, there have been initial steps toward releasing parallel benchmarks targeting the future CMPs [22].”
Pond: CXL-Based Memory Pooling Systems for Cloud Platforms (2022, neutral): “Specifically, our workloads span in-memory databases and KV-stores (Redis [22], VoltDB [27], and TPC-H on MySQL [26]), data and graph processing (Spark [19] and GAPBS [33]), HPC (SPLASH2x [81]), CPU and shared-memory benchmarks (SPEC CPU [24] and PARSEC [35]), and a range of Azure’s internal work-lo”
McPAT: An integrated power, area, and timing modeling framework for multicore and manycore architectures (2009, positive): “The same subset of PARSEC benchmark suite is used for the OOO simulations.”
Relaxing non-volatility for fast and energy-efficient STT-RAM caches (2011, methodology): “We evaluate our STT-RAM designs using three metrics: (i) speedup, (ii) total energy used by the caches, and (iii) energy-delay (E-D) product for a set of multithreaded workloads drawn from the PARSEC 2.0 benchmark suite [ 3 ], [4].”
Benchmarking TPU, GPU, and CPU Platforms for Deep Learning (2019, background): “Indeed, benchmarks have been the driving force for compiler and architecture design for decades, and notable examples include the SPEC CPU [25] and PARSEC multiprocessor benchmarks [7].”
Post-compiler software optimization for reducing energy (2014, methodology): “We use the popular PARSEC [9] benchmark suite of programs representing “emerging workloads.””
Performance Measurement : Current Perspectives and Future Challenges (2006, positive): “We tested LIME using data gathered from 15 parallel sections in multithreaded applications from SPLASH-2 [24] and PARSEC [4] benchmark suites.”
Improving GPU performance via large warps and two-level warp scheduling (2011, methodology): “We created parallel applications adapted from existing ben chmark suites including Rodinia [7], MineBench [19], PARSEC [5], and NVIDIA’s CUDA SDK code sampl es [21] in addition to creating one benchmark of our own (blackjack).”