Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “caches”

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 271 records · Page 15

MAPredict: Static Analysis Driven Memory Access Prediction Framework for Modern CPUs

Application memory access patterns are crucial in deciding how much traffic is served by the cache and forwarded to the dynamic random-access memory (DRAM). However, predicting such memory traffic is difficult because of the interplay of prefetchers, compilers, parallel execution, and innovations in manufacturer-specific micro-architectures. This research introduced MAPredict, a static analysis-driven framework that addresses these challenges to predict last-level cache (LLC)-DRAM traffic. By exploring and analyzing the behavior of modern Intel processors, MAPredict formulates cache-aware analytical models. MAPredict invokes these models to predict LLC-DRAM traffic by combining the application model, machine model, and user-provided hints to capture dynamic information. MAPredict successfully predicts LLC-DRAM traffic for different regular access patterns and provides the means to combine static and empirical observations for irregular access patterns. Evaluating 130 workloads from six applications on recent Intel micro-architectures, MAPredict yielded an average accuracy of 99% for streaming, 91% for strided, and 92% for stencil patterns. By coupling static and empirical methods, up to 97% average accuracy was obtained for random access patterns on different micro-architectures.

Monil, M. A. H.↗

Operational Experience with the Frontier System in CMS

The Frontier framework is used in the CMS experiment at the LHC to deliver conditions data to processing clients worldwide, including calibration, alignment, and configuration information. Each central server at CERN, called a Frontier Launchpad, uses tomcat as a servlet container to establish the communication between clients and the central Oracle database. HTTP-proxy Squid servers, located close to clients, cache the responses to queries in order to provide high performance data access and to reduce the load on the central Oracle database. Each Frontier Launchpad also has its own reverse-proxy Squid for caching. The three central servers have been delivering about 5 million responses every day since the LHC startup, containing about 40 GB data in total, to more than one hundred Squid servers located worldwide, with an average response time on the order of 10 milliseconds. The Squid caches deployed worldwide process many more requests per day, over 700 million, and deliver over 40 TB of data. Several monitoring tools of the tomcat log files, the accesses of the Squids on the central Launchpad servers, and the availability of remote Squids have been developed to guarantee the performance of the service and make the system easily maintainable. Following a brief introduction of the Frontier framework, we describe the performance of this highly reliable and stable system, detail monitoring concerns and their deployment, and discuss the overall operational experience from the first two years of LHC data-taking.

Blumenfeld, Barry↗

Accessing Data Federations with CVMFS

Data federations have become an increasingly common tool for large collaborations such as CMS and Atlas to efficiently distribute large data files. Unfortunately, these typically are implemented with weak namespace semantics and a non-POSIX API. On the other hand, CVMFS has provided a POSIX-compliant read-only interface for use cases with a small working set size (such as software distribution). The metadata required for the CVMFS POSIX interface is distributed through a caching hierarchy, allowing it to scale to the level of about a hundred thousand hosts. In this paper, we will describe our contributions to CVMFS that merges the data scalability of XRootD-based data federations (such as AAA) with metadata scalability and POSIX interface of CVMFS. We modified CVMFS so it can serve unmodified files without copying them to the repository server. CVMFS 2.2.0 is also able to redirect requests for data files to servers outside of the CVMFS content distribution network. Finally, we added the ability to manage authorization and authentication using security credentials such as X509 proxy certificates. We combined these modifications with the OSGs StashCache regional XRootD caching infrastructure to create a cached data distribution network. Here, we will show performance metrics accessing the data federation through CVMFS compared to direct data federation access. Additionally, we will discuss the improved user experience of providing access to a data federation through a POSIX filesystem.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Exploiting Internal Parallelism for Address Translation in Solid-State Drives

Solid-state Drives (SSDs) have changed the landscape of storage systems and present a promising storage solution for data-intensive applications due to their low latency, high bandwidth, and low power consumption compared to traditional hard disk drives. SSDs achieve these desirable characteristics using internal parallelism—parallel access to multiple internal flash memory chips—and a Flash Translation Layer (FTL) that determines where data are stored on those chips so that they do not wear out prematurely. However, current state-of-the-art cache-based FTLs like the Demand-based Flash Translation Layer (DFTL) do not allow IO schedulers to take full advantage of internal parallelism, because they impose a tight coupling between the logical-to-physical address translation and the data access. In this study to address this limitation, we introduce a new FTL design called Parallel-DFTL that works with the DFTL to decouple address translation operations from data accesses. Parallel-DFTL separates address translation and data access operations into different queues, allowing the SSD to use concurrent flash accesses for both types of operations. We also present a Parallel-LRU cache replacement algorithm to improve the concurrency of address translation operations. To compare Parallel-DFTL against existing FTL approaches, we present a Parallel-DFTL performance model and compare its predictions against those for DFTL and an ideal page-mapping approach. We also implemented the Parallel-DFTL approach in an SSD simulator using real device parameters, and used trace-driven simulation to evaluate Parallel-DFTL’s efficacy. Our evaluation results show that Parallel-DFTL improved the overall performance by up to 32% for the real IO workloads we tested, and by up to two orders of magnitude with synthetic test workloads. Finally, we also found that Parallel-DFTL is able to achieve reasonable performance with a very small cache size and that it provides the best benefit for those workloads with large request size or with high write ratio.

97 MATHEMATICS AND COMPUTING↗

MAC: Memory Access Coalescer for 3D-Stacked Memory

Emerging data-intensive applications, such as graph analytics and data mining, exhibit irregular memory access patterns. Research has shown that with these memory-bound applications, traditional cache-based processor architectures, which exploit locality and regular patterns to mitigate the memory-wall issue, are inefficient. Meantime, novel 3D-stacked memory devices, such as Hybrid Mem- ory Cube (HMC) and High Bandwidth Memory (HBM), promise significant increases in bandwidth that appear extremely appealing for memory-bound applications. However, conventional memory interfaces designed for cache-based architectures and JEDEC DDR devices fit poorly with the 3D-stacked memory, which leads to significant under-utilization of the promised high bandwidth. As a response to these issues, in this paper we propose MAC (Memory Access Coalescer), a coalescing unit for the 3D-stacked memory. We discuss the design and implementation of MAC, in the context of a custom designed cache-less architecture targeted at data-intensive, irregular applications. Through a custom simulation infrastructure based on the RISC-V toolchain, we show that MAC achieves a coalescing efficiency of 52.85% on average. It improves the performance of the memory system by 60.73% on average for a large set of irregular workloads.

Wang, Xi↗

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗

Spy in the GPU-box: Covert and Side Channel Attacks on Multi-GPU System

The deep learning revolution has been enabled in large part by GPUs, and more recently accelerators, which make it possible to carry out computationally demanding training and inference in acceptable times. As the size of machine learning networks and workloads continues to increase, multi-GPU machines have emerged as an important platform offered on High Performance Computing and cloud data centers. Since these machines are shared among multiple users, it becomes increasingly important to protect applications against potential attacks. In this paper, we explore the vulnerability of Nvidia's DGX multi-GPU machines to covert and side channel attacks. These machines consist of a number of discrete GPUs that are interconnected through a combination of custom interconnect (NVLink) and PCIe connections. We reverse engineer the interconnected cache hierarchy and show that it is possible for an attacker on one GPU to cause contention on the L2 cache of another GPU. We use this observation to first develop a covert channel attack across two GPUs, achieving the best bandwidth of around 4 MB/s. We also develop a prime and probe attack on a remote GPU allowing an attacker to recover the cache access pattern of another workload. This access pattern can be used in any number of side channel attacks: we demonstrate a proof of concept attack that fingerprints the application running on the remote GPU, with high accuracy. We also develop a proof of concept attack to extract hyperparameters of a machine learning workload. Our work establishes for the first time the vulnerability of these machines to microarchitectural attacks and can guide future research to improve their security.

Dutta, Sankha↗

Quantum Monte Carlo Endstation for Petascale Computing

The major achievements enabled by QMC Endstation grant include * Performance improvement on clusters of x86 multi-core systems, especially on Cray XT systems * New and improved methods for the wavefunction optimizations * New forms of trial wavefunctions * Implementation of the full application on NVIDIA GPUs using CUDA The scaling studies of QMCPACK on large-scale systems show excellent parallel efficiency up to 216K cores on Jaguarpf (Cray XT5). The GPU implementation shows speedups of 10-15x over the CPU implementation on older generation of x86. We have implemented hybrid OpenMP/MPI scheme in QMC to take advantage of multi-core shared memory processors of petascale systems. Our hybrid scheme has several advantages over the standard MPI-only scheme. * Memory optimized: large read-only data to store one-body orbitals and other shared properties to represent the trial wave function and many-body Hamiltonian can be shared among threads, which reduces the memory footprint of a large-scale problem. * Cache optimized: the data associated with an active Walker are in cache during the compute-intensive drift-diffusion process and the operations on an Walker are optimized for cache reuse. Thread-local objects are used to ensure the data affinity to a thread. * Load balanced: Walkers in an ensemble are evenly distributed among threads and MPI tasks. The two-level parallelism reduces the population imbalance among MPI tasks and reduces the number of point-to-point communications of large messages (serialized objects) for the Walker exchange. * Communication optimized: the communication overhead, especially for the collective operations necessary to determine ET and measure the properties of an ensemble, is significantly lowered by using less MPI tasks. The multiple forms of parallelism afforded by QMC algorithms make them ideal candidates for acceleration in the many-core paradigm. We presented the results of our effort to port the QMCPACK simulation code to the NVIDIA CUDA GPU platform. We restructured the CPU algorithms to express additional parallelism, minimize GPU-CPU communication, and efficiently utilize the GPU memory hierarchy. Using mixed precision on GT200 GPUs and MPI for intercommunication and load balancing, we observe typical full-application speedups of approximately 10x to 15x relative to quad-core Xeon CPUs alone, while reproducing the double-precision CPU results within statistical error. We developed an all-electron quantum Monte Carlo (QMC) method for solids that does not rely on pseudopotentials, and used it to construct a primary ultra-high-pressure calibration based on the equation of state of cubic boron nitride. We computed the static contribution to the free energy with the QMC method and obtained the phonon contribution from density functional theory, yielding a high-accuracy calibration up to 900 GPa usable directly in experiment. We computed the anharmonic Raman frequency shift with QMC simulations as a function of pressure and temperature, allowing optical pressure calibration. In contrast to present experimental approaches, small systematic errors in the theoretical EOS do not increase with pressure, and no extrapolation is needed. This all-electron method is applicable to first-row solids, providing a new reference for ab initio calculations of solids and benchmarks for pseudopotential accuracy. We compared experimental and theoretical results on the momentum distribution and the quasiparticle renormalization factor in sodium. From an x-ray Compton-profile measurement of the valence-electron momentum density, we derived its discontinuity at the Fermi wavevector finding an accurate measure of the renormalization factor that we compared with quantum-Monte-Carlo and G0W0 calculations performed both on crystalline sodium and on the homogeneous electron gas. Our calculated results are in good agreement with the experiment. We have been studying the heat of formation for various Kubas complexes of molecular hydrogen on Ti(1,2)ethylene-nH2 using Diffusion Monte Carlo. This work has been started and is ongoing. We are studying systems involving 1 and 2 Ti bonding sites with up to 10 hydrogen molecules in numerous configurations. This work will establish a benchmark that will test the accuracy of density functional calculations and establish the feasibility of our methods for similar systems.

97 MATHEMATICS AND COMPUTING↗

Sparse Matrix-Matrix Multiplication on Multilevel Memory Architectures: Algorithms and Experiments

Architectures with multiple classes of memory media are becoming a common part of mainstream supercomputer deployments. So called multi-level memories offer differing characteristics for each memory component including variation in bandwidth, latency and capacity. This paper investigates the performance of sparse matrix multiplication kernels on two leading highperformance computing architectures — Intel's Knights Landing processor and NVIDIA's Pascal GPU. We describe a data placement method and a chunking-based algorithm for our kernels that exploits the existence of the multiple memory spaces in each hardware platform. We evaluate the performance of these methods w.r.t. standard algorithms using the auto-caching mechanisms Our results show that standard algorithms that exploit cache reuse performed as well as multi-memory-aware algorithms for architectures such as Ki\iLs where the memory subsystems have similar latencies. However, for architectures such as GPUS where memory subsystems differ significantly in both bandwidth and latency, multi-memory-aware methods are crucial for good performance. In addition, our new approaches permit the user to run problems that require larger capacities than the fastest memory of each compute node without depending on the software-managed cache mechanisms.

97 MATHEMATICS AND COMPUTING↗

An Integrated Framework for Memory-Centric Analysis: From Trace Collection to Co-Design

The memory wall phenomenon—where advances in processor performance significantly outpace those in memory subsystems—poses a fundamental challenge for contemporary computing systems. In memory-bound applications, memory subsystem behavior dominates performance, yet existing analysis approaches present significant limitations: detailed microarchitectural simulators require days to weeks to simulate modest workloads; hardware performance counters provide only aggregate statistics that obscure temporal and spatial access patterns; and scaled simulation approaches face challenges in capturing certain behaviors that emerge at larger scales. These limitations reflect a processor-centric design philosophy increasingly misaligned with memory-bound workloads where detailed understanding of memory access patterns, cache hierarchy interactions, and contention is critical for effective optimization. This paper presents an integrated framework for memory-centric analysis that enables effective hardware-software co-design. We describe practical trace collection techniques, including hardware-assisted processor tracing with minimal overhead and portable software-based instrumentation with statistical sampling. We present multi-perspective analysis methods that examine memory behavior from temporal, sequential, spatial, and relational viewpoints, revealing distinct optimization opportunities invisible in aggregate metrics. We detail an architectural modeling framework that uses sampled traces with temporal interpolation and confidence-based filtering to evaluate cache and memory configurations. Evaluation on representative benchmarks demonstrates that this framework achieves practical accuracy (L2 cache errors of 2.64\%, confidence-filtered L3 errors of 9.92\%, bandwidth errors of 7.33\%) while providing substantial speedup (26.8×) over cycle-accurate simulation, enabling rapid design space exploration. We demonstrate how this integrated framework enables systematic identification of both hardware optimizations (memory controller tuning, bank partitioning, NUMA configuration) and software optimizations (data layout restructuring, prefetching strategies, memory-aware scheduling). Through this comprehensive treatment of the memory-centric analysis pipeline—from trace collection through architectural modeling to co-design application—we provide researchers and practitioners with practical techniques for addressing memory bottlenecks in contemporary computing systems.

Gajaria, Dhruv Mayur↗

Is random access memory random?

Most software is contructed on the assumption that the programs and data are stored in random access memory (RAM). Physical limitations on the relative speeds of processor and memory elements lead to a variety of memory organizations that match processor addressing rate with memory service rate. These include interleaved and cached memory. A very high fraction of a processor's address requests can be satified from the cache without reference to the main memory. The cache requests information from main memory in blocks that can be transferred at the full memory speed. Programmers who organize algorithms for locality can realize the highest performance from these computers.

Denning, P. J.↗

Prefetching in file systems for MIMD multiprocessors

The question of whether prefetching blocks on the file into the block cache can effectively reduce overall execution time of a parallel computation, even under favorable assumptions, is considered. Experiments have been conducted with an interleaved file system testbed on the Butterfly Plus multiprocessor. Results of these experiments suggest that (1) the hit ratio, the accepted measure in traditional caching studies, may not be an adequate measure of performance when the workload consists of parallel computations and parallel file access patterns, (2) caching with prefetching can significantly improve the hit ratio and the average time to perform an I/O (input/output) operation, and (3) an improvement in overall execution time has been observed in most cases. In spite of these gains, prefetching sometimes results in increased execution times (a negative result, given the optimistic nature of the study). The authors explore why it is not trivial to translate savings on individual I/O requests into consistently better overall performance and identify the key problems that need to be addressed in order to improve the potential of prefetching techniques in the environment.

Kotz, David F.↗

Empirical study of parallel LRU simulation algorithms

This paper reports on the performance of five parallel algorithms for simulating a fully associative cache operating under the LRU (Least-Recently-Used) replacement policy. Three of the algorithms are SIMD, and are implemented on the MasPar MP-2 architecture. Two other algorithms are parallelizations of an efficient serial algorithm on the Intel Paragon. One SIMD algorithm is quite simple, but its cost is linear in the cache size. The two other SIMD algorithm are more complex, but have costs that are independent on the cache size. Both the second and third SIMD algorithms compute all stack distances; the second SIMD algorithm is completely general, whereas the third SIMD algorithm presumes and takes advantage of bounds on the range of reference tags. Both MIMD algorithm implemented on the Paragon are general and compute all stack distances; they differ in one step that may affect their respective scalability. We assess the strengths and weaknesses of these algorithms as a function of problem size and characteristics, and compare their performance on traces derived from execution of three SPEC benchmark programs.

Carr, Eric↗

Becoming Reactive by Concretization

One way to build a reactive system is to construct an action table indexed by the current situation or stimulus. The action table describes what course of action to pursue for each situation or stimulus. This paper describes an incremental approach to constructing the action table through achieving goals with a hierarchical search system. These hierarchies are generated with transformations called concretizations, which add constraints to a problem and which can reduce the search space. The basic idea is that an action for a state is looked up in the action table and executed whenever the action table has an entry for that state; otherwise, a path is found to the nearest (cost-wise in a graph with costweighted arcs) state that has a mappring from a state in the next highest hierarchy. For each state along the solution path, the successor state in the path is cached in the action table entry for that state. Without caching, the hierarchical search system can logarithmically reduce search. When the table is complete the system no longer searches: it simply reacts by proceeding to the state listed in the table for each state. Since the cached information is specific only to the nearest state in the next highest hierarchy and not the goal, inter-goal transfer of reactivity is possible. To illustrate our approach, we show how an implemented hierarchical search system can completely reactive.

Prieditis, Armand↗

The Science of Computing: Virtual Memory

In the March-April issue, I described how a computer's storage system is organized as a hierarchy consisting of cache, main memory, and secondary memory (e.g., disk). The cache and main memory form a subsystem that functions like main memory but attains speeds approaching cache. What happens if a program and its data are too large for the main memory? This is not a frivolous question. Every generation of computer users has been frustrated by insufficient memory. A new line of computers may have sufficient storage for the computations of its predecessor, but new programs will soon exhaust its capacity. In 1960, a longrange planning committee at MIT dared to dream of a computer with 1 million words of main memory. In 1985, the Cray-2 was delivered with 256 million words. Computational physicists dream of computers with 1 billion words. Computer architects have done an outstanding job of enlarging main memories yet they have never kept up with demand. Only the shortsighted believe they can.

Denning, Peter J.↗

ROPEC - ROtary PErcussive Coring Drill for Mars Sample Return

The ROtary Percussive Coring Drill is a light weight, flight-like, five-actuator drilling system prototype designed to acquire core material from rock targets for the purposes of Mars Sample Return. In addition to producing rock cores for sample caching, the ROPEC drill can be integrated with a number of end effectors to perform functions such as rock surface abrasion, dust and debris removal, powder and regolith acquisition, and viewing of potential cores prior to caching. The ROPEC drill and its suite of end effectors have been demonstrated with a five degree of freedom Robotic Arm mounted to a mobility system with a prototype sample cache and bit storage station.

Chu, Philip↗

X-Ray Computed Tomography: The First Step in Mars Sample Return Processing

The Mars 2020 rover mission will collect and cache samples from the martian surface for possible retrieval and subsequent return to Earth. If the samples are returned, that mission would likely present an opportunity to analyze returned Mars samples within a geologic context on Mars. In addition, it may provide definitive information about the existence of past or present life on Mars. Mars sample return presents unique challenges for the collection, containment, transport, curation and processing of samples [1] Foremost in the processing of returned samples are the closely paired considerations of life detection and Planetary Protection. In order to achieve Mars Sample Return (MSR) science goals, reliable analyses will depend on overcoming some challenging signal/noise-related issues where sparse martian organic compounds must be reliably analyzed against the contamination background. While reliable analyses will depend on initial clean acquisition and robust documentation of all aspects of developing and managing the cache [2], there needs to be a reliable sample handling and analysis procedure that accounts for a variety of materials which may or may not contain evidence of past or present martian life. A recent report [3] suggests that a defined set of measurements should be made to effectively inform both science and Planetary Protection, when applied in the context of the two competing null hypotheses: 1) that there is no detectable life in the samples; or 2) that there is martian life in the samples. The defined measurements would include a phased approach that would be accepted by the community to preserve the bulk of the material, but provide unambiguous science data that can be used and interpreted by various disciplines. Fore-most is the concern that the initial steps would ensure the pristine nature of the samples. Preliminary, non-invasive techniques such as computed X-ray tomography (XCT) have been suggested as the first method to interrogate and characterize the cached samples without altering the materials [1,2]. A recent report [4] indicates that XCT may minimally alter samples for some techniques, and work is needed to quantify these effects, maximizing science return from XCT initial analysis while minimizing effects.

Welzenbach, L. C.↗

Organic and inorganic contamination control approaches for return sample investigation on Mars 2020

The Mars 2020 Rover mission will have the capability to collect and cache samples for potential Mars sample return. Specifically, the sample caching system (SCS) is designed for coring Mars samples and acquiring regolith samples as well as handling, sealing and caching on Mars. As the potential first Martian samples that could be returned to Earth, assuring low levels of terrestrial contamination is of the utmost concern. In developing the SCS, the project prioritizes limiting sample contamination in organic, inorganic and biological areas. The focus of this paper is on the strategies being implemented to limit terrestrial organic and inorganic contamination in the samples.

Steltzner, Adam↗