Tools for Power-Energy Modelling and Analysis of Parallel Scientific Applications

Alonso, Pedro; Badía, Rosa M.; Labarta, Jesús; Barreda, Maria; Dolz, Manuel F.; Mayo, Rafael; Quintana‐Ortí, Enrique S.; Reyes, Ruyman

doi:10.1109/icpp.2012.57

Cited by 56 publications

(39 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the ODROID-XU3 board the pmlib monitoring tool [30] collects power consumption corresponding to instantaneous power readings from four independent sensors/power domains in the board (Cortex-A7 cluster, Cortex-A15 cluster, DRAM and GPU), with a sampling rate of 250 ms. (To compensate for this low sampling rate, our calibration experiments repeat the execution of the kernels for a period that is sufficiently long to obtain enough power measurements.) When evaluating the energy of one of the clusters, we only consider the sensor corresponding to that component.…”

Section: Methodsmentioning

confidence: 99%

Energy balance between voltage-frequency scaling and resilience for linear algebra routines on low-power multicore architectures

et al. 2018

View full text Add to dashboard Cite

Near Threshold Voltage (NTV) computing has been recently proposed as a technique to save energy, at the cost of incurring higher error rates including, among others, Silent Data Corruption (SDC). In this paper, we evaluate the energy efficiency of dense linear algebra routines using several low-power multicore processors and we analyze whether the potential energy reduction achieved when scaling the processor to operate at a low voltage compensates the cost of integrating a fault tolerance mechanism that tackles SDC. Our study targets algorithmic-based fault-tolerant versions of the dense matrix-vector and matrix(-matrix) multiplication kernels (gemv and gemm, respectively), using the BLIS framework, as well as an implementation of the LU factorization with partial pivoting built on top of gemm. Furthermore, we tailor the study for a number of representative 32-bit and 64-bit multicore processors from ARM that were specifically designed for energy efficiency.

show abstract

Section: Methodsmentioning

confidence: 99%

Energy balance between voltage-frequency scaling and resilience for linear algebra routines on low-power multicore architectures

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Score-P v3.1 (Knüpfer et al, 2012) and Scalasca v2.3.1 (Geimer et al, 2010;Zhukov et al, 2015), where results collected with Score-P and Scalasca can be examined using the interactive analysis report explorer Cube v4.3.5 , Allinea Performance Reports v7.0.4 (January et al, 2015), Extrae v3.4.3 (Alonso et al, 2012), Paraver v4.6.3 (Labarta et al, 2006), Intel Vectorization Advisor 2015 (Rane et al, 2015), and Darshan v3.0.0 (Carns et al, 2011) (see Table A1 10 in Appendix A for a more detailed description of each performance analysis tool listed above). The modeling chain for the profiling workflow is as follows:…”

Section: Code Profilingmentioning

confidence: 99%

Best practice regarding the three P's: profiling, portability and provenance when running HPC geoscientific applications

Sharples

Zhukov

Geimer

et al. 2017

Preprint

View full text Add to dashboard Cite

Abstract. Geoscientific modeling is constantly evolving, with next generation geoscientific models and applications placing high demands on high performance computing (HPC) resources. These demands are being met by new developments in HPC architectures, software libraries, and infrastructures. New HPC developments require new programming paradigms leading to substantial investment in model porting, tuning, and refactoring of complicated legacy code in order to use these resources effectively. In addition to the challenge of new massively parallel HPC systems, reproducibility of simulation and analysis 5 results is of great concern, as the next generation geoscientific models are based on complex model implementations and profiling, modeling and data processing workflows.Thus, in order to reduce both the duration and the cost of code migration, aid in the development of new models or model components, while ensuring reproducibility and sustainability over the complete data life cycle, a streamlined approach to profiling, porting, and provenance tracking is necessary. We propose a run control framework (RCF) integrated with a workflow 10 engine which encompasses all stages of the modeling chain: 1. preprocess input, 2. compilation of code (including code instrumentation with performance analysis tools), 3. simulation run, 4. postprocess and analysis, to address these issues. Within this RCF, the workflow engine is used to create and manage benchmark or simulation parameter combinations and performs the documentation and data organization for reproducibility. This approach automates the process of porting and tuning, profiling, testing, and running a geoscientific model. We show that in using our run control framework, testing, benchmarking, profiling, 15 and running models is less time consuming and more robust, resulting in more efficient use of HPC resources, more strategic code development, and enhanced data integrity and reproducibility.

show abstract

“…The wattmeter is connected via an Ethernet link to a separate power tracing server that runs a daemon application to collect power samples form the internal wattmeter. The measurement application is built by calling routines from the pmlib library [8,13]. We set the sampling rate to 1 kSamples/sec., which is high enough to obtain reliable measures for the power model and remaining experiments.…”

Section: Environment Setupmentioning

confidence: 99%

Assessing the impact of the CPU power-saving modes on the task-parallel solution of sparse linear systems

et al. 2014

Self Cite

View full text Add to dashboard Cite

We investigate the benefits that an energyaware implementation of the runtime in charge of the concurrent execution of ILUPACK -a sophisticated preconditioned iterative solver for sparse linear systems-produces on the time-power-energy balance of the application. Furthermore, to connect the experimental results with the theory, we propose several simple yet accurate power models that capture the variations of average power that result from the introduction of the energy-aware strategies as well as the impact of the P-states into ILUPACK's runtime, at high accuracy, on two distinct platforms based on multicore technology from AMD and Intel.

show abstract

Tools for Power-Energy Modelling and Analysis of Parallel Scientific Applications

Cited by 56 publications

References 8 publications

Energy balance between voltage-frequency scaling and resilience for linear algebra routines on low-power multicore architectures

Energy balance between voltage-frequency scaling and resilience for linear algebra routines on low-power multicore architectures

Best practice regarding the three P's: profiling, portability and provenance when running HPC geoscientific applications

Assessing the impact of the CPU power-saving modes on the task-parallel solution of sparse linear systems

Contact Info

Product

Resources

About