Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userPerformance Modeling and Prediction for Dense Linear Algebra
61
0
0.0
(
0
)
Added by
Elmar Peise
Publication date
2017
fields
Informatics Engineering
and research's language is
English
Authors
Elmar Peise
Ask ChatGPT about the research
No Arabic abstract
This dissertation introduces measurement-based performance modeling and prediction techniques for dense linear algebra algorithms. As a core principle, these techniques avoid executions of such algorithms entirely, and instead predict their performance through runtime estimates for the underlying compute kernels. For a variety of operations, these predictions allow to quickly select the fastest algorithm configurations from available alternatives. We consider two scenarios that cover a wide range of computations: To predict the performance of blocked algorithms, we design algorithm-independent performance models for kernel operations that are generated automatically once per platform. For various matrix operations, instantaneous predictions based on such models both accurately identify the fastest algorithm, and select a near-optimal block size. For performance predictions of BLAS-based tensor contractions, we propose cache-aware micro-benchmarks that take advantage of the highly regular structure inherent to contraction algorithms. At merely a fraction of a contractions runtime, predictions based on such micro-benchmarks identify the fastest combination of tensor traversal and compute kernel.
rate research
Read More
Information-driven networks include a large category of networking systems, where network nodes are aware of information delivered and thus can not only forward data packets but may also perform information processing. In many situations, the quality of service (QoS) in information-driven networks is provisioned with the redundancy in information. Traditional performance models generally adopt evaluation measures suitable for packet-oriented service guarantee, such as packet delay, throughput, and packet loss rate. These performance measures, however, do not align well with the actual need of information-driven networks. New performance measures and models for information-driven networks, despite their importance, have been mainly blank, largely because information processing is clearly application dependent and cannot be easily captured within a generic framework. To fill the vacancy, we present a new performance evaluation framework particularly tailored for information-driven networks, based on the recent development of stochastic network calculus. We analyze the QoS with respect to information delivery and study the scheduling problem with the new performance metrics. Our analytical framework can be used to calculate the network capacity in information delivery and in the meantime to help transmission scheduling for a large body of systems where QoS is stochastically guaranteed with the redundancy in information.
The life-cycle of a partial differential equation (PDE) solver is often characterized by three development phases: the development of a stable numerical discretization, development of a correct (verified) implementation, and the optimization of the implementation for different computer architectures. Often it is only after significant time and effort has been invested that the performance bottlenecks of a PDE solver are fully understood, and the precise details varies between different computer architectures. One way to mitigate this issue is to establish a reliable performance model that allows a numerical analyst to make reliable predictions of how well a numerical method would perform on a given computer architecture, before embarking upon potentially long and expensive implementation and optimization phases. The availability of a reliable performance model also saves developer effort as it both informs the developer on what kind of optimisations are beneficial, and when the maximum expected performance has been reached and optimisation work should stop. We show how discretization of a wave equation can be theoretically studied to understand the performance limitations of the method on modern computer architectures. We focus on the roofline model, now broadly used in the high-performance computing community, which considers the achievable performance in terms of the peak memory bandwidth and peak floating point performance of a computer with respect to algorithmic choices. A first principles analysis of operational intensity for key time-stepping finite-difference algorithms is presented. With this information available at the time of algorithm design, the expected performance on target computer systems can be used as a driver for algorithm design.
The A64FX CPU is arguably the most powerful Arm-based processor design to date. Although it is a traditional cache-based multicore processor, its peak performance and memory bandwidth rival accelerator devices. A good understanding of its performance features is of paramount importance for developers who wish to leverage its full potential. We present an architectural analysis of the A64FX used in the Fujitsu FX1000 supercomputer at a level of detail that allows for the construction of Execution-Cache-Memory (ECM) performance models for steady-state loops. In the process we identify architectural peculiarities that point to viable generic optimization strategies. After validating the model using simple streaming loops we apply the insight gained to sparse matrix-vector multiplication (SpMV) and the domain wall (DW) kernel from quantum chromodynamics (QCD). For SpMV we show why the CRS matrix storage format is not a good practical choice on this architecture and how the SELL-C-sigma format can achieve bandwidth saturation. For the DW kernel we provide a cache-reuse analysis and show how an appropriate choice of data layout for complex arrays can realize memory-bandwidth saturation in this case as well. A comparison with state-of-the-art high-end Intel Cascade Lake AP and Nvidia V100 systems puts the capabilities of the A64FX into perspective. We also explore the potential for power optimizations using the tuning knobs provided by the Fugaku system, achieving energy savings of about 31% for SpMV and 18% for DW.
Networks-on-Chip (NoCs) used in commercial many-core processors typically incorporate priority arbitration. Moreover, they experience bursty traffic due to application workloads. However, most state-of-the-art NoC analytical performance analysis techniques assume fair arbitration and simple traffic models. To address these limitations, we propose an analytical modeling technique for priority-aware NoCs under bursty traffic. Experimental evaluations with synthetic and bursty traffic show that the proposed approach has less than 10% modeling error with respect to cycle-accurate NoC simulator.
The A64FX CPU powers the current number one supercomputer on the Top500 list. Although it is a traditional cache-based multicore processor, its peak performance and memory bandwidth rival accelerator devices. Generating efficient code for such a new architecture requires a good understanding of its performance features. Using these features, we construct the Execution-Cache-Memory (ECM) performance model for the A64FX processor in the FX700 supercomputer and validate it using streaming loops. We also identify architectural peculiarities and derive optimization hints. Applying the ECM model to sparse matrix-vector multiplication (SpMV), we motivate why the CRS matrix storage format is inappropriate and how the SELL-C-sigma format with suitable code optimizations can achieve bandwidth saturation for SpMV.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
