Search NASA⌕ Search

SEARCH · Search NASA

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

Spin-transfer torque magnetoresistive random-access memory technologies for normally off computing (invited)

Most parts of present computer systems are made of volatile devices, and the power to supply them to avoid information loss causes huge energy losses. We can eliminate this meaningless energy loss by utilizing the non-volatile function of advanced spin-transfer torque magnetoresistive random-access memory (STT-MRAM) technology and create a new type of computer, i.e., normally off computers. Critical tasks to achieve normally off computers are implementations of STT-MRAM technologies in the main memory and low-level cache memories. STT-MRAM technology for applications to the main memory has been successfully developed by using perpendicular STT-MRAMs, and faster STT-MRAM technologies for applications to the cache memory are now being developed. The present status of STT-MRAMs and challenges that remain for normally off computers are discussed.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Scaling to 150K cores: recent algorithm and performance engineering developments enabling XGC1 to run at scale

Particle-in-cell (PIC) methods have proven to be effective in discretizing the Vlasov-Maxwell system of equations describing the core of toroidal burning plasmas for many decades. Recent physical understanding of the importance of edge physics for stability and transport in tokamaks has lead to development of the fi rst fully toroidal edge PIC code - XGC1. The edge region poses special problems in meshing for PIC methods due to the lack of closed flux surfaces, which makes fi eld-line following meshes and coordinate systems problematic. We present a solution to this problem with a semi- field line following mesh method in a cylindrical coordinate system. Additionally, modern supercomputers require highly concurrent algorithms and implementations, with all levels of the memory hierarchy being effe ciently utilized to realize optimal code performance. This paper presents a mesh and particle partitioning method, suitable to our meshing strategy, for use on highly concurrent cache-based computing platforms.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The 'last mile' of data handling: Fermilab's IFDH tools

IFDH (Intensity Frontier Data Handling), is a suite of tools for data movement tasks for Fermilab experiments and is an important part of the FIFE[2] (Fabric for Intensity Frontier [1] Experiments) initiative described at this conference. IFDH encompasses moving input data from caches or storage elements to compute nodes (the 'last mile' of data movement) and moving output data potentially to those caches as part of the journey back to the user. IFDH also involves throttling and locking to ensure that large numbers of jobs do not cause data movement bottlenecks. IFDH is realized as an easy to use layer that users call in their job scripts (e.g. 'ifdh cp'), hiding the low level data movement tools. One advantage of this layer is that the underlying low level tools can be selected or changed without the need for the user to alter their scripts. Logging and performance monitoring can also be added easily. This system will be presented in detail as well as its impact on the ease of data handling at Fermilab experiments.

Lyon, Adam L.↗

Virtual machine provisioning, code management, and data movement design for the Fermilab HEPCloud Facility

The Fermilab HEPCloud Facility Project has as its goal to extend the current Fermilab facility interface to provide transparent access to disparate resources including commercial and community clouds, grid federations, and HPC centers. This facility enables experiments to perform the full spectrum of computing tasks, including data-intensive simulation and reconstruction. We have evaluated the use of the commercial cloud to provide elasticity to respond to peaks of demand without overprovisioning local resources. Full scale data-intensive workflows have been successfully completed on Amazon Web Services for two High Energy Physics Experiments, CMS and NOνA, at the scale of 58000 simultaneous cores. This paper describes the significant improvements that were made to the virtual machine provisioning system, code caching system, and data movement system to accomplish this work. The virtual image provisioning and contextualization service was extended to multiple AWS regions, and to support experiment-specific data configurations. A prototype Decision Engine was written to determine the optimal availability zone and instance type to run on, minimizing cost and job interruptions. We have deployed a scalable on-demand caching service to deliver code and database information to jobs running on the commercial cloud. It uses the frontiersquid server and CERN VM File System (CVMFS) clients on EC2 instances and utilizes various services provided by AWS to build the infrastructure (stack). We discuss the architecture and load testing benchmarks on the squid servers. We also describe various approaches that were evaluated to transport experimental data to and from the cloud, and the optimal solutions that were used for the bulk of the data transport. Finally, we summarize lessons learned from this scale test, and our future plans to expand and improve the Fermilab HEP Cloud Facility.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Scidac-Data: Enabling Data Driven Modeling of Exascale Computing

Here, the SciDAC-Data project is a DOE-funded initiative to analyze and exploit two decades of information and analytics that have been collected by the Fermilab data center on the organization, movement, and consumption of high energy physics (HEP) data. The project analyzes the analysis patterns and data organization that have been used by NOvA, MicroBooNE, MINERvA, CDF, D0, and other experiments to develop realistic models of HEP analysis workflows and data processing. The SciDAC-Data project aims to provide both realistic input vectors and corresponding output data that can be used to optimize and validate simulations of HEP analysis. These simulations are designed to address questions of data handling, cache optimization, and workflow structures that are the prerequisites for modern HEP analysis chains to be mapped and optimized to run on the next generation of leadership-class exascale computing facilities. We present the use of a subset of the SciDAC-Data distributions, acquired from analysis of approximately 71,000 HEP workflows run on the Fermilab data center and corresponding to over 9 million individual analysis jobs, as the input to detailed queuing simulations that model the expected data consumption and caching behaviors of the work running in high performance computing (HPC) and high throughput computing (HTC) environments. In particular we describe how the Sequential Access via Metadata (SAM) data-handling system in combination with the dCache/Enstore-based data archive facilities has been used to develop radically different models for analyzing the HEP data. We also show how the simulations may be used to assess the impact of design choices in archive facilities.

97 MATHEMATICS AND COMPUTING↗

A Flexible-blocking Based Approach for Performance Tuning of Matrix Multiplication Routines for Large Matrices with Edge Cases

Efficient and scalable matrix operations are being highly demanding in the recent era of Machine Learning, Deep Learning, and Big Data Analytics. The two commonly used matrix-matrix operations in the Basic Linear Algebra Subprograms (BLAS) specification are General Matrix-Matrix multiplication (GEMM) and Symmetric Rank-k update (SYRK). The SYRK routine is a specialization of the GEMM routine, where half of the multiplications are skipped as the resultant matrix is known to be symmetric. Fortunately, several linear algebra libraries implement these BLAS routines quite efficiently. The libraries usually partition the input matrices into blocks and place them in processor caches, thus improving performance by leveraging the caches. However, the contemporary libraries are highly optimized for squarish matrices, but the performance degrades significantly for the matrices with edge case (strictly thin or strictly fat shapes) in the multicore machine. The primary reason is that the current state-of-the-art libraries make fixed block shapes based on a processor architecture, and do not consider the shape of the input matrices. In this paper, we propose a new blocking approach, we name it Flexible-blocking, to mitigate the scalability issues. In contrast to the contemporary libraries, our approach formulates the blocks of the input matrices based on the shapes of the matrices as well as the number of threads used in the implementation. Our proposed technique shows noticeable performance improvement on multicore shared-memory machines for the edge case matrices.

Hossain, Md Mosharaf↗

Locality Aware Concurrent Start for Stencil Applications

Stencil computations are at the heart of many physical simulations used in scientific codes. Thus, there exists a plethora of optimization efforts for this family of computations. Among these techniques, tiling techniques that allow concurrent start have proven to be very efficient in providing better performance for these critical kernels. Nevertheless, with many core designs being the norm, these optimization techniques might not be able to fully exploit locality (both spatial and temporal) on multiple levels of the memory hierarchy without compromising parallelism. It is no longer true that the machine can be seen as a homogeneous collection of nodes with caches, main memory and an interconnect network. New architectural designs exhibit complex grouping of nodes, cores, threads, caches and memory connected by an ever evolving network-on-chip design. These new designs may benefit greatly from carefully crafted schedules and groupings that encourage parallel actors (i.e. threads, cores or nodes) to be aware of the computational history of other actors in close proximity. In this paper, we provide an efficient tiling technique that allows hierarchical concurrent start for memory hierarchy aware tile groups. Each execution schedule and tile shape exploit the available parallelism, load balance and locality present in the given applications. We demonstrate our technique on the Intel Xeon Phi architecture with selected and representative stencil kernels. We show improvement ranging from 5.58% to 31.17% over existing state-of-the-art techniques.

Locality Aware execution, jagged tiling↗

Understanding Strong Scaling on GPUs Using Empirical Performance Saturation Size

The roofline model provides a concise overview of the maximum performance capabilities of a given computer system through a combination of peak memory bandwidth and compute performance rates. The increasing complexity of scheduling and cache in recent GPUs, however, has introduced complicated performance variability that is not captured by arithmetic intensity alone. This work examines the effect of problem size and GPU launch configurations on roofline performance for V100, A100, MI100, and MI250X graphics processing units. We introduce an extended roofline model that takes problem size into account, and find that strong scaling on GPUs can be characterized by saturation problem sizes as additional key metrics. Saturation problem sizes break up a plot of GPU performance vs. problem size into three distinct performance regimes– size-limited, cache-bound, and DRAM-bound. With our extended roofline model, we are able to provide a robust view of these performance regimes across recent GPU architectures.

Eberius, David↗

Shifting Between Compute and Memory Bounds: A Compression-Enabled Roofline Model

In the evolving landscape of high-performance computing, especially to fight the end of Moore’s Law and Dennard’s Scaling, the ability to shift between compute-bound and memory-bound states is critical for enhancing adaptability and flexibility to diverse system and domain-specific architectures. Such capability is vital for optimizing performance across distinguished hardware configurations, such as accelerators, memory hierarchies, and cache systems. Despite that ad hoc optimization techniques, such as compressed/approximate computation, have been enabled for compute-/data-intensive computing for improved performance in distinct hardware settings, there lacks an understanding of 1) the rational behind performance improvement; 2) capability of different optimizations; 3) what optimization to respond to specific computational and memory demands. This work proposes a compression-enabled roofline model to facilitate this adaptability with data compression techniques to balance and transform between computational and memory demands. This model enables applications to adjust in response to the specific strengths and limitations of the underlying hardware and system to optimize resource utilization. The effectiveness of this approach is demonstrated with matrix multiplication kernels on different input sizes, with turning on/off various compression techniques, including 1) low-precision floating point; 2) sparse matrix formulation; and 3) compressed arrays with ZFP. By reducing memory transfer volumes and cache misses and increasing data locality and computational intensity through compression, the specific roofline model can transform between compute and memory bounds to align more efficiently with system capabilities. This advancement not only improves overall performance but also maximizes adaptability in diverse computing environments.

Naraparaju, Ramasoumya [University of Washington]↗

FPDeep: Scalable Acceleration of CNN Training on Deeply-Pipelined FPGA Clusters

In this paper, we propose a framework called FPDeep, which uses a hybrid of model and layer paral- lelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation.

Wang, Tianqi↗

A survey of techniques for architecting and managing GPU register file

To support their massively-multithreaded architecture, GPUs use very large register file (RF) which has a capacity higher than even L1 and L2 caches. In total contrast, traditional CPUs use tiny RF and much larger caches to optimize latency. Due to these differences, along with the crucial impact of RF in determining GPU performance, novel and intelligent techniques are required for managing GPU RF. In this paper, we survey the techniques for designing and managing GPU RF. We discuss techniques related to performance, energy and reliability aspects of RF. To emphasize the similarities and differences between the techniques, we classify them along several parameters. Lastly, the aim of this paper is to synthesize the state-of-art developments in RF management and also stimulate further research in this area.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Locality-Aware CTA Clustering For Modern GPUs

In this paper, we proposed a novel clustering technique for tapping into the performance potential of a largely ignored type of locality: inter-CTA locality. We first demonstrated the capability of the existing GPU hardware to exploit such locality, both spatially and temporally, on L1 or L1/Tex unified cache. To verify the potential of this locality, we quantified its existence in a broad spectrum of applications and discussed its sources of origin. Based on these insights, we proposed the concept of CTA-Clustering and its associated software techniques. Finally, We evaluated these techniques on all modern generations of NVIDIA GPU architectures. The experimental results showed that our proposed clustering techniques could significantly improve on-chip cache performance.

Data Locality↗

Moment representation in the lattice Boltzmann method on massively parallel hardware

The widely-used lattice Boltzmann method (LBM) for computational fluid dynamics is highly scalable, but also significantly memory bandwidth-bound on current architectures. This paper presents a new regularized LBM implementation that reduces the memory footprint by only storing macroscopic, moment-based data. We show that the amount of data that must be stored in memory during a simulation is reduced by up to 47%. We also present a technique for cache-aware data re-utilization and show that optimizing cache utilization to limit data motion results in a similar improvement in time to solution. These new algorithms are implemented in the hemodynamics solver HARVEY and demonstrated using both idealized and realistic biological geometries. We develop a performance model for the moment representation algorithm and evaluate the performance on Summit.

Vardhan, Madhurima↗

Design and analysis of CXL performance models for tightly-coupled heterogeneous computing

Truly heterogeneous systems enable partitioned workloads to be mapped to the hardware that nets the best performance. However, current practice requires that inter-device communication between different vendors' hardware use host memory as an intermediary step. To date, there are no widely adopted solutions that allow accelerators to directly transfer data. A new cache-coherent protocol, CXL, aims to facilitate easier, fine-grained sharing between accelerators. In this work we analyze existing methods for designing heterogeneous applications that target GPUs and FPGAs working collaboratively, followed by an exploration to show the benefits of a CXL-enabled system. Specifically, we develop a test application that utilizes both an NVIDIA P100 GPU and a Xilinx U250 FPGA to show current communication limitations. From this application, we capture overall execution time and throughput measurements on the FPGA and GPU. We use these measurements as inputs to novel CXL performance models to show that using CXL caching instead of host memory results in a 1.31X speedup, while a more tightly-coupled pipelined implementation using CXL-enabled hardware would result in a speedup of 1.45X.

Cabrera, Anthony↗

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗

Performance Impact and Trade-Offs for Tuning Key Architectural Parameters on CPU+GPU Systems

In this work, we performed an initial design space exploration of an accelerated processing unit (APU)—a hybrid CPU+GPU architecture that integrates both compute units (CUs) and memory into a unified system. This integration aims to reduce data movement, enhance memory locality, and improve energy efficiency by enabling the CPU and GPU to share memory directly. This effort focused on the interplay of key design components—cache line size, the number of CUs, and main memory technology—and the trade-offs of each configuration were analyzed. This paper highlights the various configurations’ impact on memory accesses, data reuse, and power utilization. The results provide valuable insights that can be leveraged to optimize APU architectures for high-performance and energy-efficient computing and thus create a balanced architecture. This optimization can be achieved by adopting dynamic cache management, runtime CU scaling, and advanced memory integration, highlighting the potential of APUs to address critical challenges in compute, data movement, and memory power consumption.

Asifuzzaman, Kazi [ORNL] (ORCID:0000000240044791)↗

T-FSM: A Scalable Distributed Task-Based System for Frequent Subgraph Pattern Mining from a Big Graph

Finding frequent subgraph patterns in a big graph is an important problem with many applications such as classifying chemical compounds and building indexes to speed up graph queries. Since this problem is NP-hard, some recent parallel and distributed systems have been developed to accelerate the mining. However, they often have a huge memory cost, very long running time, suboptimal load balancing, poor scale-out capability, and possibly inaccurate results. In this article, we propose an efficient system called T-FSM for parallel mining of frequent subgraph patterns in a big graph. T-FSM supports a new anti-monotonic frequentness measure called Fraction-Score, which is more accurate than the widely used MNI measure. The execution engine of T-FSM supports both intra-machine parallelism and inter-machine parallelism. For intra-machine parallelism, T-FSM adopts a novel task-based execution model to ensure high multithreading concurrency, bounded memory consumption, and effective load balancing. For inter-machine parallelism, T-FSM ensures good scale-out performance with a lightweight pattern rebalancing approach that reduces workload skewness of pattern evaluations among machines. To avoid recomputing the contexts for migrated patterns, we design a novel context cache table to support concurrent and asynchronous requesting and caching of remote context data, which can timely evict and garbage collect used pattern contexts that are no longer needed to keep memory consumption bounded. Extensive experiments show that T-FSM is orders of magnitude faster than existing state-of-the-art parallel systems (more than 10×, 51×, 131×, 55× speedup over ScaleMine, DistGraph, Pangolin and Peregrine, respectively) and distributed systems (more than 42× and 88× over ScaleMine and DistGraph, respectively) for frequent subgraph pattern mining, and it scales out satisfactorily to 512 CPU cores on the Polaris supercomputer at Argonne National Laboratory.

97 MATHEMATICS AND COMPUTING↗