Search

Many High Performance Computing (HPC) facilities have developed and deployed frameworks in support of continuous monitoring and operational data analytics (MODA) to help improve efficiency and throughput. Because of the complexity and scale of systems and workflows and the need for low-latency response to address dynamic circumstances, automated feedback and response have the potential to be more effective than current human-in-the-loop approaches which are laborious and error prone. Progress has been limited, however, by factors such as the lack of infrastructure and feedback hooks, and successful deployment is often site- and case-specific. In this position paper we report on the outcomes and plans from a recent Dagstuhl Seminar, seeking to carve a path for community progress in the development of autonomous feedback loops for MODA, based on the established formalism of similar (MAPE-K) loops in autonomous computing and self-adaptive systems. By defining and developing such loops for significant cases experienced across HPC sites, we seek to extract commonalities and develop conventions that will facilitate interoperability and interchangeability with system hardware, software, and applications across different sites, and will motivate vendors and others to provide telemetry interfaces and feedback hooks to enable community development and pervasive deployment of MODA autonomy loops.

This paper addresses the challenge of providing portable and highly efficient code structures for CPU and GPU architectures. We choose the assembly of the right-hand term in the incompressible flow module of the High-Performance Computational Mechanics code Alya, which is one of the two CFD codes in the Unified European Benchmark Suite. Starting from an efficient CPU-code and a related OpenACC-port for GPUs we successively investigate performance potentials arising from code specialization, algorithmic restructuring and low-level optimizations. We demonstrate that only the combination of these different dimensions of runtime optimization unveils the full performance potential on the GPU and CPU. Roofline-based performance modelling is applied in this process and we demonstrate the need to investigate new optimization strategies if a classical roofline limit such as memory bandwidth utilization is achieved, rather than stopping the process. The final unified OpenACC-based implementation boosts performance by more than 50x on an NVIDIA A100 GPU (achieving approximately 2.5 TF/s FP64) and a further factor of 5x for an Intel Icelake based CPU-node (achieving approximately 1.0 TF/s FP64). The insights gained in our manual approach lays ground implementing unified but still highly efficient code structures for related kernels in Alya and other applications. These can be realized by manual coding or automatic code generation frameworks.

In this paper we analyze the MPI-only version of the CloverLeaf code from the SPEChpc 2021 benchmark suite on recent Intel Xeon "Ice Lake" and "Sapphire Rapids" server CPUs. We observe peculiar breakdowns in performance when the number of processes is prime. Investigating this effect, we create first-principles data traffic models for each of the stencil-like hotspot loops. With application measurements and microbenchmarks to study memory data traffic behavior, we can connect the breakdowns to SpecI2M, a new write-allocate evasion feature in current Intel CPUs. For serial and full-node cases we are able to predict the memory data volume analytically with an error of a few percent. We find that if the number of processes is prime, SpecI2M fails to work properly, which we can attribute to short inner loops emerging from the one-dimensional domain decomposition in this case. We can also rule out other possible causes of the prime number effect, such as breaking layer conditions, MPI communication overhead, and load imbalance., Comment: 19 pages including artifact appendix; 11 figures, 1 table; numerous corrections, esp. in Table 1

Long term observation of temporal Earth's gravity field with enhanced temporal and spatial resolution is a major objective for future satellite gravity missions. Improving the performance of the accelerometers present in such missions is one of the main path to explore. In this context, we propose to study an original concept of a hybrid accelerometer associating a state-of-the-art electrostatic accelerometer (EA) and a promising quantum sensor based on cold atom interferometry. To assess the performance potential of such instrument, numerical simulations have been performed to determine its impact in term of gravity field retrieval. Taking advantage of the long term stability of the cold atom interferometer (CAI), it has been shown that the reduced drift of the hybrid sensor could lead to improved gravity field retrieval. Nevertheless this gain vanishes once temporal variations of the gravity field and related aliasing effects are taken into account. Improved de-aliasing models or some specific satellite constellations are then required to maximize the impact of the accelerometer performance gain. To evaluate the achievable acceleration performance in orbit, a numerical simulator of the hybrid accelerometer has been developed and preliminary results are given. The instrument simulator has been in part validated by reproducing the performance achieved with a hybrid lab prototype operating on ground. The problematic of satellite rotation impact on the CAI has been also investigated both with instrument performance simulations and experimental demonstrations. It has been shown that the proposed configuration, where the EA's proof-mass acts as the reference mirror for the CAI, seems a promising approach to allow the mitigation of satellite rotation. A preliminary design has been elaborated along with a preliminary error, mass, volume and electrical power consumption budget., Comment: 47 pages, 31 figures

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., Comment: 32 pages, 25 figures, 6 tables

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., Comment: 6 pages, 5 figures, 3 tables

Modern electronic structure theories can predict and simulate a wealth of phenomena in surface science and solid-state physics. In order to allow for a direct comparison with experiment, such ab initio predictions have to be made in the thermodynamic limit, substantially increasing the computational cost of many-electron wave-function theories. Here, we present a method that achieves thermodynamic limit results for solids and surfaces using the "gold standard" coupled cluster ansatz of quantum chemistry with unprecedented efficiency. We study the energy difference between carbon diamond and graphite crystals, adsorption energies of water on h-BN, as well as the cohesive energy of the Ne solid, demonstrating the increased efficiency and accuracy of coupled cluster theory for solids and surfaces., Comment: 4 figures

Stencil algorithms have been receiving considerable interest in HPC research for decades. The techniques used to approach multi-core stencil performance modeling and engineering span basic runtime measurements, elaborate performance models, detailed hardware counter analysis, and thorough scaling behavior evaluation. Due to the plurality of approaches and stencil patterns, we set out to develop a generalizable methodology for reproducible measurements accompanied by state-of-the-art performance models. Our open-source toolchain, and collected results are publicly available in the "Intranode Stencil Performance Evaluation Collection" (INSPECT). We present the underlying methodologies, models and tools involved in gathering and documenting the performance behavior of a collection of typical stencil patterns across multiple architectures and hardware configuration options. Our aim is to endow performance-aware application developers with reproducible baseline performance data and validated models to initiate a well-defined process of performance assessment and optimization.

We present an ab-initio study of boron nitride as well as carbon allotropes. Their relative thermodynamic stabilities and structural phase transitions from low- to high-density phases are investigated. Pressure-temperature phase diagrams are calculated and compared to experimental findings. The calculations are performed using quantum chemical wavefunction based as well as density functional theories. Our findings reveal that predicted energy differences often depend significantly on the choice of the employed method. Comparison between calculated and experimental results allows for benchmarking the accuracy of various levels of theory. The produced results show that quantum chemical wavefunction based theories allow for achieving systematically improvable estimates. We find that on the level of coupled cluster theories the low- and high-density phases of boron nitride become thermodynamically degenerate at 0 K. This is in agreement with recent experimental findings, indicating that cubic boron nitride is not the thermodynamically stable allotrope at ambient conditions. Furthermore we employ the calculated results to assess transition probabilities from graphitic low-density to diamond-like high-density phases in an approximate manner. We conclude that the stacking order of the parent graphitic material is crucial for the possible formation of meta-stable wurtzite boron nitride and hexagonal carbon diamond also known as lonsdaleite.

We investigate crystalline thermodynamic stable lithium silicides phases (LixSiy) with density functional theory (DFT) and a force-field method based on modified embedded atoms (MEAM) and compare our results with experimental data. This work presents a fast and accurate framework to calculate thermodynamic properties of crystal structures with large unit cells with MEAM based on molecular dynamics (MD). Mechanical properties like the bulk modulus and the elastic constants are evaluated in addition to thermodynamic properties including the phonon density of states, the vibrational free energy and the isochoric/isobaric specific heat capacity for Li, Li12Si7, Li7Si3, Li13Si4, Li15Si4, Li21Si5, Li17Si4, Li22Si5 and Si. For a selected phase (Li13Si4) we study the effect of a temperature dependent phonon density of states and its effect on the isobaric heat capacity.