Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “cache”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Leaky Buddies: Cross-Component Covert Channels on Integrated CPU-GPU Systems

Graphics Processing Units (GPUs) are ubiquitous components used across the range of today’s computing plat- forms, from phones and tablets, through personal computers, to high-end server class platforms. With the increasing importance of graphics and video workloads, recent processors are shipped with GPU devices that are integrated on the same chip. Integrated GPUs share some resources with the CPU and as a result, there is a potential for microarchitectural attacks from the GPU to the CPU or vice versa. We consider the potential for covert channel attacks that arise either from shared microarchitectural components (such as caches) or through shared contention domains (e.g., shared buses). We illustrate these two types of channels by developing two reliable covert channel attacks. The first covert channel uses the shared LLC cache in Intel’s integrated GPU architectures. The second is a contention based channel targeting the ring bus connecting the CPU and GPU to the LLC. This is the first demonstrated microarchitectural attack crossing the component boundary (GPU to CPU or vice versa). Cross-component channels introduce a number of new challenges that we had to overcome since they occur across heterogeneous components that use different computation models and are interconnected using asymmetric memory hierarchies. We also exploit GPU parallelism to increase the bandwidth of the communication, even without relying on a common clock. The LLC based channel achieves a bandwidth of 120 kbps with a low error rate of 2%, while the contention based channel delivers up to 400 kbps with a 0.8% error rate. We also demonstrate a proof- of-concept prime-and-probe side channel attack that probes the full LLC from the GPU.

Cyber secruity, high performance comptuing, Hetero↗
DOE OSTI2022

Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks

This work develops new techniques within Horovod, a generic communication library supporting data parallel training across deep learning frameworks. In particular, we improve the Horovod control plane by implementing a new coordination scheme that takes advantage of the characteristics of the typical data parallel training paradigm, namely the repeated execution of collectives on the gradients of a fixed set of tensors. Using a caching strategy, we execute Horovod’s existing coordinator-worker logic only once during a typical training run, replacing it with a more efficient decentralized orchestration strategy using the cached data and a global intersection of a bitvector for the remaining training duration. Next, we introduce a feature for end users to explicitly group collective operations, enabling finer grained control over the communication buffer sizes. To evaluate our proposed strategies, we conduct experiments on a world-class supercomputer — Summit. We compare our proposals to Horovod’s original design and observe 2x performance improvement at a scale of 6000 GPUs; we also compare them against tf.distribute and torch.DDP and achieve 12% better and comparable performance, respectively, using up to 1536 GPUs; we compare our solution against BytePS in typical HPC settings and achieve about 20% better performance on a scale of 768 GPUs. Finally, we test our strategies on a scientific application (STEMDL) using up to 27,600 GPUs (the entire Summit) and show that we achieve a near-linear scaling of 0.93 with a sustained performance of 1.54 exaflops (with standard error +- 0.02) in FP16 precision.

Romero, Joshua↗
DOE OSTI2022

SLIP (Surrogate Launching and Integration Platform)

SLIP (Surrogate Launching and Integration Platform) is a software ecosystem for downloading and running ML/AI benchmarks. SLIP automatically downloads and sets up the code and data running a benchmark. The large data sets and model code are cached locally with a specific ID within a designated cache directory.

Brown, Cade↗
DOE OSTI2016

An efficient and portable SIMD algorithm for charge/current deposition in Particle-In-Cell codes

In current computer architectures, data movement (from die to network) is by far the most energy consuming part of an algorithm (≈20pJ/word on-die to ≈10,000 pJ/word on the network). To increase memory locality at the hardware level and reduce energy consumption related to data movement, future exascale computers tend to use many-core processors on each compute nodes that will have a reduced clock speed to allow for efficient cooling. To compensate for frequency decrease, machine vendors are making use of long SIMD instruction registers that are able to process multiple data with one arithmetic operator in one clock cycle. SIMD register length is expected to double every four years. As a consequence, Particle-In-Cell (PIC) codes will have to achieve good vectorization to fully take advantage of these upcoming architectures. In this paper, we present a new algorithm that allows for efficient and portable SIMD vectorization of current/charge deposition routines that are, along with the field gathering routines, among the most time consuming parts of the PIC algorithm. Our new algorithm uses a particular data structure that takes into account memory alignment constraints and avoids gather/scat;ter instructions that can significantly affect vectorization performances on current CPUs. The new algorithm was successfully implemented in the 3D skeleton PIC code PICSAR and tested on Haswell Xeon processors (AVX2-256 bits wide data registers). Results show a factor of ×2 to ×2.5 speed-up in double precision for particle shape factor of orders 1–3. The new algorithm can be applied as is on future KNL (Knights Landing) architectures that will include AVX-512 instruction sets with 512 bits register lengths (8 doubles/16 singles). Program summary Program Title: vec_deposition Program Files doi:http://dx.doi.org/10.17632/nh77fv9k8c.1 Licensing provisions: BSD 3-Clause Programming language: Fortran 90 External routines/libraries: OpenMP > 4.0 Nature of problem: Exascale architectures will have many-core processors per node with long vector data registers capable of performing one single instruction on multiple data during one clock cycle. Data register lengths are expected to double every four years and this pushes for new portable solutions for efficiently vectorizing Particle-In-Cell codes on these future many-core architectures. One of the main hotspot routines of the PIC algorithm is the current/charge deposition for which there is no efficient and portable vector algorithm. Solution method: Here we provide an efficient and portable vector algorithm of current/charge deposition routines that uses a new data structure, which significantly reduces gather/scatter operations. Vectorization is controlled using OpenMP 4.0 compiler directives for vectorization which ensures portability across different architectures. Restrictions: Here we do not provide the full PIC algorithm with an executable but only vector routines for current/charge deposition. These scalar/vector routines can be used as library routines in your 3D Particle-In-Cell code. However, to get the best performances out of vector routines you have to satisfy the two following requirements: (1) Your code should implement particle tiling (as explained in the manuscript) to allow for maximized cache reuse and reduce memory accesses that can hinder vector performances. The routines can be used directly on each particle tile. (2) You should compile your code with a Fortran 90 compiler (e.g Intel, gnu or cray) and provide proper alignment flags and compiler alignment directives (more details in README file).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Amorphous Indium Oxide Channel FEFETs With Write Voltage of 0.9 V and Endurance >10 12 for Refresh-Free Embedded Memory

This work presents, for the first time, a back-end-of-the-line (BEOL)-compatible W-doped indium oxide (IWO) ferroelectric field-effect transistor (FEFET) with a record-low operating voltage below 0.9 V and a write speed of 20 ns while achieving a transient read current window (CW) ratio ( I LVT /I HVT ) greater than 10 4 . The device also exhibits exceptional reliability characteristics such as: 1) measured bipolar write endurance up to 10 12 cycles; 2) a fast read speed of 50 ns; 3) read endurance surpassing 10 12 cycles; and 4) retention exceeding 10 4 s at 85 ∘ C. Furthermore, a physics-based numerical model has been developed to investigate the nanoscale characteristics of BEOL FEFET devices, leveraging nucleation-limited switching in HfO 2 ferroelectrics and dc characterization to extract material and channel parameters for accurate device simulation. The simulation uncovers the stochastic switching behavior of BEOL amorphous oxide semiconductor (AOS) FEFETs and demonstrates an intrinsic switching time as low as 1 ps, highlighting the potential of BEOL AOS FEFETs for ultrafast memory applications. These results establish AOS FEFETs as a compelling candidate for high-density embedded memory applications for last-level cache (LLC) (L4) in advanced CMOS technology nodes.

1-V ferroelectric field-effect transistor (FEFET)↗
DOE OSTI2020

Quantum Dynamics at Scale: Ultrafast Control of Emergent Functional Materials

Confluence of extreme-scale quantum dynamics simulations (i.e. quantum@scale) and cutting-edge x-ray free-electron laser experiments are revolutionizing materials science. An archetypal example is the exciting concept of using picosecond light pulses to control emergent material properties on demand in atomically-thin layered materials. This paper describes efforts to scale our quantum molecular dynamics engine toward the United States' first exaflop/s computer, under an Aurora Early Science Program project named "Metascalable layered material genome". Key algorithmic and computing techniques incorporated are: (1) globally-scalable and locally-fast solvers within a linear-scaling divide-conquer-recombine algorithmic framework; (2) algebraic `BLASification' of computational kernels; and (3) data alignment and loop restructuring, along with register and cache blocking, for enhanced vectorization and efficient memory access. The resulting weak-scaling parallel efficiency was 0.93 on 131,072 Intel Xeon Phi cores for a 56.6 million atom (or 169 million valence-electron) system, whereas the various code transformations achieved 5-fold speedup. The optimized simulation engine allowed us for the first time to establish a significant effect of substrate on the dynamics of layered material upon electronic excitation.

Algebraic BLASification↗
DOE OSTI2022

Performance assessment of ensembles of in situ workflows under resource constraints

Summary Scientific breakthroughs in biomolecular methods and improvements in hardware technology have shifted from a long‐running simulation to a large set of shorter simulations running simultaneously, called an ensemble. In an ensemble, simulations are usually coupled with analyses of data produced by the simulations. In situ methods can be used to analyze large volumes of data generated by scientific simulations at runtime (i.e., simulations and analyses are performed concurrently). In this work, we study the execution of ensemble‐based simulations paired with in situ analyses using in‐memory staging methods. Using an ensemble of molecular dynamics in situ workflows with multiple simulations and analyses, we first show that collecting traditional metrics such as makespan, instructions per cycle, memory usage, or cache miss ratio is not sufficient to characterize complex behaviors of ensembles. We propose a method to evaluate the performance of ensembles of workflows that captures multiple resource usage aspects: resource efficiency, resource allocation, and resource provisioning. Experimental results demonstrate that the proposed method can effectively distinguish the performance of different component placements in an ensemble with up to 32 ensemble members. By evaluating different co‐location scenarios, our proposed performance indicators demonstrate benefits of co‐locating simulation and coupled analyses within a compute node.

Do, Tu Mai Anh↗
DOE OSTI2022

Just-in-Time Compilation and Link-Time Optimization for OpenMP Target Offloading

Following the mass adoption of external accelerators for high performance computing, the overall performance of many applications has become increasingly dependent on relatively small accelerated kernels. As static analysis is fundamentally limited by dynamic values and external definitions, standard ahead-of-time compilation is not always sufficient to achieve the best performance. Furthermore, many users looking to port an existing application to run on an external accelerator will not want to fundamentally restructure their programs. These and other problems can be addressed through both link-time optimization (LTO) and just-in-time (JIT) compilation, but until now had sparse and inconsistent support from the compiler. In this work, we present a new compilation method that enables device-side LTO as well as a transparent JIT compilation tool-chain for OpenMP target offloading. Our contributions include an entirely new device linking and embedding scheme to enable LTO as well as a novel JIT engine to efficiently optimize OpenMP offloading regions at run-time. We also introduce a persistent caching system to improve end-to-end runtime using the JIT engine and minimize kernel launching overheads. We measure the performance of our LTO and JIT implementation via several real-world scientific applications. With our optimizations we observe significant improvements through LTO on large applications as well as significant end-to-end execution time improvement using JIT.

Tian, Shilei↗
DOE OSTI2025

Benchmarking Operators in Deep Neural Networks for Improving Performance Portability of SYCL

SYCL is a portable programming model for heterogeneous computing, so it is important to obtain reasonable performance portability of SYCL. Towards the goal of better understanding and improving performance portability of SYCL for machine learning workloads, we have been developing benchmarks for basic operators in deep neural networks (DNNs). These operators could be offloaded to heterogeneous computing devices such as graphics processing units (GPUs) to speed up computation. In this paper, we introduce the benchmarks, evaluate the performance of the operators on GPU-based systems, and describe the causes of the performance gap between the SYCL and Compute Unified Device Architecture (CUDA) kernels. We find that the causes are related to the utilization of the texture cache for read-only data, optimization of the memory accesses with strength reduction, use of local memory, and register usage per thread. We hope that the efforts of developing benchmarks for studying performance portability will stimulate discussion and interactions within the community.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗
DOE OSTI2016

Optimizing Excited-State Electronic-Structure Codes for Intel Knights Landing: A Case Study on the BerkeleyGW Software

We profile and optimize calculations performed with the BerkeleyGW code on the Xeon-Phi architecture. BerkeleyGW depends both on hand-tuned critical kernels as well as on BLAS and FFT libraries. We describe the optimization process and performance improvements achieved. We discuss a layered parallelization strategy to take advantage of vector, thread and node-level parallelism. We discuss locality changes (including the consequence of the lack of L3 cache) and effective use of the on-package high-bandwidth memory. We show preliminary results on Knights-Landing including a roofline study of code performance before and after a number of optimizations. We find that the GW method is particularly well-suited for many-core architectures due to the ability to exploit a large amount of parallelism over plane-wave components, band-pairs, and frequencies.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Delaunay walk for fast nearest neighbor: accelerating correspondence matching for ICP

Point set registration algorithms such as Iterative Closest Point (ICP) are commonly utilized in time-constrained environments like robotics. Finding the nearest neighbor of a point in a reference 3D point set is a common operation in ICP and frequently consumes at least 90% of the computation time. We introduce a novel approach to performing the distance-based nearest neighbor step based on Delaunay triangulation. This greedy algorithm finds the nearest neighbor of a query point by traversing the edges of the Delaunay triangulation created from a reference 3D point set. Our work integrates the Delaunay traversal into the correspondences search of ICP and exploits the iterative aspect of ICP by caching previous correspondences to expedite each iteration. An algorithmic analysis and comparison is conducted showing an order of magnitude speedup for both serial and vector processor implementation.

3d point cloud processing↗
DOE OSTI2021

Perseverance’s Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals (SHERLOC) Investigation

The Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals (SHERLOC) is a robotic arm-mounted instrument on NASA’s Perseverance rover. SHERLOC has two primary boresights. The Spectroscopy boresight generates spatially resolved chemical maps using fluorescence and Raman spectroscopy coupled to microscopic images (10.1 μm/pixel). The second boresight is a Wide Angle Topographic Sensor for Operations and eNgineering (WATSON); a copy of the Mars Science Laboratory (MSL) Mars Hand Lens Imager (MAHLI) that obtains color images from microscopic scales (~13 μm/pixel) to infinity. SHERLOC Spectroscopy focuses a 40 μs pulsed deep UV neon-copper laser (248.6 nm), to a ~100 μm spot on a target at a working distance of ~48 mm. Fluorescence emissions from organics, and Raman scattered photons from organics and minerals, are spectrally resolved with a single diffractive grating spectrograph with a spectral range of 250 to ~370 nm. Because the fluorescence and Raman regions are naturally separated with deep UV excitation (<250 nm), the Raman region ~ 800 – 4000 cm -1 (250 to 273 nm) and the fluorescence region (274 to ~370 nm) are acquired simultaneously without time gating or additional mechanisms. SHERLOC science begins by using an Autofocus Context Imager (ACI) to obtain target focus and acquire 10.1 μm/pixel greyscale images. Chemical maps of organic and mineral signatures are acquired by the orchestration of an internal scanning mirror that moves the focused laser spot across discrete points on the target surface where spectra are captured on the spectrometer detector. ACI images and chemical maps (< 100 μm/mapping pixel) will enable the first Mars in situ view of the spatial distribution and interaction between organics, minerals, and chemicals important to the assessment of potential biogenicity (containing CHNOPS). Single robotic arm placement chemical maps can cover areas up to 7x7 mm in area and, with the < 10 min acquisition time per map, larger mosaics are possible with arm movements. This microscopic view of the organic geochemistry of a target at the Perseverance field site, when combined with the other instruments, such as Mastcam-Z, PIXL, and SuperCam, will enable unprecedented analysis of geological materials for both scientific research and determination of which samples to collect and cache for Mars sample return.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Persistent Sampling: Enhancing the Efficiency of Sequential Monte Carlo

Sequential Monte Carlo (SMC) samplers are powerful tools for Bayesian inference but suffer from high computational costs due to their reliance on large particle ensembles for accurate estimates. We introduce persistent sampling (PS), an extension of SMC that systematically retains and reuses particles from all prior iterations to construct a growing, weighted ensemble. By leveraging multiple importance sampling and resampling from a mixture of historical distributions, PS mitigates the need for excessively large particle counts, directly addressing key limitations of SMC such as particle impoverishment and mode collapse. Crucially, PS achieves this without additional likelihood evaluations-weights for persistent particles are computed using cached likelihood values. This framework not only yields more accurate posterior approximations but also produces marginal likelihood estimates with significantly lower variance, enhancing reliability in model comparison. Furthermore, the persistent ensemble enables efficient adaptation of transition kernels by leveraging a larger, decorrelated particle pool. Experiments on high-dimensional Gaussian mixtures, hierarchical models, and non-convex targets demonstrate that PS consistently outperforms standard SMC and related variants, including recycled and waste-free SMC, achieving substantial reductions in mean squared error for posterior expectations and evidence estimates, all at reduced computational cost. PS thus establishes itself as a robust, scalable, and efficient alternative for complex Bayesian inference tasks.

Karamanis, Minas↗
DOE OSTI2014

HybridPlan: A Capacity Planning Technique for Projecting Storage Requirements in Hybrid Storage Systems

Economic forces, driven by the desire to introduce flash into the high-end storage market without changing existing software-base, have resulted in the emergence of solid-state drives (SSDs), flash packaged in HDD form factors and capable of working with device drivers and I/O buses designed for HDDs. Unlike the use of DRAM for caching or buffering, however, certain idiosyncrasies of NAND Flash-based solid-state drives (SSDs) make their integration into hard disk drive (HDD)-based storage systems nontrivial. Flash memory suffers from limits on its reliability, is an order of magnitude more expensive than the magnetic hard disk drives (HDDs), and can sometimes be as slow as the HDD (due to excessive garbage collection (GC) induced by high intensity of random writes). Given the complementary properties of HDDs and SSDs in terms of cost, performance, and lifetime, the current consensus among several storage experts is to view SSDs not as a replacement for HDD, but rather as a complementary device within the high-performance storage hierarchy. Thus, we design and evaluate such a hybrid storage system with HybridPlan that is an improved capacity planning technique to administrators with the overall goal of operating within cost-budgets. HybridPlan is able to find the most cost-effective hybrid storage configuration with different types of SSDs and HDDs

Capacity Planning↗
DOE OSTI2021

Tonal Emergence: An agent-based model of tonal coordination

Humans have a remarkable capacity for coordination. Our ability to interact and act jointly in groups is crucial to our success as a species. Joint Action (JA) research has often concerned itself with simplistic behaviors in highly constrained laboratory tasks. But there has been a growing interest in understanding complex coordination in more open-ended contexts. In this regard, collective music improvisation has emerged as a fascinating model domain for studying basic JA mechanisms in an unconstrained and highly sophisticated setting. A number of empirical studies have begun to elucidate coordination mechanisms underlying joint musical improvisation, but these empirical findings have yet to be cached out in a working computational model. The present work fills this gap by presenting TonalEmergence, an idealized agent-based model of improvised musical coordination. TonalEmergence models the coordination of notes played by improvisers to generate harmony (i.e., tonality), by simulating agents that stochastically generate notes biased towards maximizing harmonic consonance given their partner’s previous notes. Here, the model replicates an interesting empirical result from a previous study of professional jazz pianists: feedback loops of mutual adaptation between interacting agents support the production of consonant harmony. The model is further explored to show how complex tonal dynamics, such as the production and dissolution of stable tonal centers, are supported by agents that are characterized by (i) a tendency to strive toward consonance, (ii) stochasticity, and (iii) a limited memory for previously played notes. TonalEmergence thus provides a grounded computational model to simulate and probe the coordination mechanisms underpinning one of the more remarkable feats of human cognition: collective music improvisation.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

An efficient tensor transpose algorithm for multicore CPU, Intel Xeon Phi, and NVidia Tesla GPU

An efficient parallel tensor transpose algorithm is suggested for shared-memory computing units, namely, multicore CPU, Intel Xeon Phi, and NVidia GPU. The algorithm operates on dense tensors (multidimensional arrays) and is based on the optimization of cache utilization on x86 CPU and the use of shared memory on NVidia GPU. From the applied side, the ultimate goal is to minimize the overhead encountered in the transformation of tensor contractions into matrix multiplications in computer implementations of advanced methods of quantum many-body theory (e.g., in electronic structure theory and nuclear physics). A particular accent is made on higher-dimensional tensors that typically appear in the so-called multireference correlated methods of electronic structure theory. Depending on tensor dimensionality, the presented optimized algorithms can achieve an order of magnitude speedup on x86 CPUs and 2-3 times speedup on NVidia Tesla K20X GPU with respect to the na ve scattering algorithm (no memory access optimization). Furthermore, the tensor transpose routines developed in this work have been incorporated into a general-purpose tensor algebra library (TAL-SH).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Pushing memory bandwidth limitations through efficient implementations of Block-Krylov space solvers on GPUs

Lattice quantum chromodynamics simulations in nuclear physics have benefited from a tremendous number of algorithmic advances such as multigrid and eigenvector deflation. These improve the time to solution but do not alleviate the intrinsic memory-bandwidth constraints of the matrix-vector operation dominating iterative solvers. Batching this operation for multiple vectors and exploiting cache and register blocking can yield a super-linear speed up. Block-Krylov solvers can naturally take advantage of such batched matrix-vector operations, further reducing the iterations to solution by sharing the Krylov space between solves. However, practical implementations typically suffer from the quadratic scaling in the number of vector-vector operations. Here, using the QUDA library, we present an implementation of a block-CG solver on NVIDIA GPUs which reduces the memory-bandwidth complexity of vector-vector operations from quadratic to linear. We present results for the HISQ discretization, showing a 5x speedup compared to highly-optimized independent Krylov solves on NVIDIA's SaturnV cluster.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2018

Large-scale GW calculations on pre-exascale HPC systems

The ab initio GW approach is a rigorous Green’s-function-based framework that can be employed to compute electronic excitation properties of a wide variety of materials such as extended systems, molecules, as well as confined and nanostructured materials with a very satisfactory accuracy. However, GW calculations on complex systems are often hindered by the high computational cost associated with the method. Here, in this study, we demonstrate how to significantly speedup GW calculations with a novel algorithm for the computationally intense kernel based on a non-blocking chunked cyclic communication scheme that minimizes latency in MPI messages, allows for overlapping of communication and computation, and improves cache usage in matrix multiplication operations. The optimized version of the code, implemented in the BerkeleyGW software package, is capable of scaling well to the full Cori computer at NERSC (Cray XC40) and achieves over 11 Peta FLOP/s of sustained performance. We showcase our work by performing large-scale GW calculations of defect structures on silicon, which require simulation cells containing over 1700 atoms, and which can now be efficiently executed in just a few minutes on large pre-exascale high-performance computing systems.

97 MATHEMATICS AND COMPUTING↗