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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Reducing the stochasticity of crystal nucleation to enable subnanosecond memory writing

Operation speed is a key challenge in phase-change random-access memory (PCRAM) technology, especially for achieving subnanosecond high-speed cache memory. Commercialized PCRAM products are limited by the tens of nanoseconds writing speed, originating from the stochastic crystal nucleation during the crystallization of amorphous germanium antimony telluride (Ge 2 Sb 2 Te 5 ). Here, we demonstrate an alloying strategy to speed up the crystallization kinetics. The scandium antimony telluride (Sc 0.2 Sb 2 Te 3 ) compound that we designed allows a writing speed of only 700 picoseconds without preprogramming in a large conventional PCRAM device. This ultrafast crystallization stems from the reduced stochasticity of nucleation through geometrically matched and robust scandium telluride (ScTe) chemical bonds that stabilize crystal precursors in the amorphous state. Controlling nucleation through alloy design paves the way for the development of cache-type PCRAM technology to boost the working efficiency of computing systems.

Science & Technology - Other Topics↗
DOE OSTI2017

On the arithmetic intensity of high-order finite-volume discretizations for hyperbolic systems of conservation laws

It has been conjectured that higher-order discretizations for partial differential equations will have advantages over the lower-order counterparts commonly used today. The reasoning is that the increase in arithmetic operations will be more than offset by the reduction in data transfers and the increase in concurrent floating-point units. To evaluate this conjecture, the arithmetic intensity of a class of high-order finite-volume discretizations for hyperbolic systems of conservation laws is theoretically analyzed for spatial discretizations from orders three through eight in arbitrary dimensions. Additionally, three cache models are considered: the limiting cases of no cache and infinite cache as well as a finite-sized cache model. Models are validated experimentally by measuring floating-point operations and data transfers on an IBM Blue Gene/Q node. Theory and experiments demonstrate that high-order finite-volume methods will be able to provide increases in arithmetic intensity that will be necessary to make better utilization of on-node floating-point capability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Evaluate data lake design for the accelerator control system

Increasing precision in automation for modern particle accelerators not only creates a requirement to gather data from all devices but also demands scalable and high-performance data infrastructure with the capability of handling vast incoming device data. A well architected data lake is suitable for such a system which integrates real-time data acquisition, transient data caching, and long-term storage. This paper evaluates data lake architecture for an Accelerator Control System (ACS), focusing on two critical components of a data lake, data cache and long-term storage.

Jaikar, Amol [Fermilab]↗
DOE OSTI2014

Blackcomb: Hardware-Software Co-design for Non-Volatile Memory in Exascale Systems

Summary of technical results of Blackcomb Memory Devices We explored various different memory technologies (STTRAM, PCRAM, FeRAM, and ReRAM). The progress can be classified into three categories, below. Modeling and Tool Releases Various modeling tools have been developed over the last decade to help in the design of SRAM or DRAM-based memory hierarchies. To explore new design opportunities that NVM technologies can bring to the designers, we have developed similar high-level models for NVM, including PCRAMsim [Dong 2009], NVSim [Dong 2012], and NVMain [Poremba 2012]. NVSim is a circuit-level model for NVM performance, energy, and area estimation, which supports various NVM technologies, including STT-RAM, PCRAM, ReRAM, and legacy NAND Flash. NVSim is successfully validated against industrial NVM prototypes, and it is expected to help boost architecture-level NVM-related studies. On the other side, NVMain is a cycle accurate main memory simulator designed to simulate emerging nonvolatile memories at the architectural level. We have released these models as open source tools and provided contiguous support to them. We also proposed PS3-RAM, which is a fast, portable and scalable statistical STT-RAM reliability analysis model [Wen 2012]. Design Space Exploration and Optimization With the support of these models, we explore different device/circuit optimization techniques. For example, in [Niu 2012a] we studied the power reduction technique for the application of ECC scheme in ReRAM designs and proposed to use ECC code to relax the BER (Bit Error Rate) requirement of a single memory to improve the write energy consumption and latency for both 1T1R and cross-point ReRAM designs. In [Xu 2011], we proposed a methodology to design STT-RAM for different optimization goals such as read performance, write performance and write energy by leveraging the trade-off between write current and write time of MTJ. We also studied the tradeoffs in building a reliable crosspoint ReRAM array [Niu 2012b]. We have conducted an in depth analysis of the circuit and system level design implications of multi-layer cross-point Resistive RAM (MLCReRAM) from performance, power and reliability perspectives [Xu 2013]. The objective of this study is to understand the design trade-offs of this technology with respect to the MLC Phase Change Memory (MLCPCM).Our MLC ReRAM design at the circuit and system levels indicates that different resistance allocation schemes, programming strategies, peripheral designs, and material selections profoundly affect the area, latency, power, and reliability of MLC ReRAM. Based on this analysis, we conduct two case studies: first we compare MLC ReRAM design against MLC phase-change memory (PCM) and multi-layer cross-point ReRAM design, and point out why multi-level ReRAM is appealing; second we further explore the design space for MLC ReRAM. Architecture and Application We explored hybrid checkpointing using phase-change memory for future exascale systems [Dong 2011] and showed that the use of nonvolatile memory for local checkpointing significantly increases the number of faults covered by local checkpoints and reduces the probability of a global failure in the middle of a global checkpoint to less than 1%. We also proposed a technique called i2WAP to mitigate the write variations in NVM-based last-level cache for the improvement of the NVM lifetime [Wang 2013]. Our wear leveling technique attempts to work around the limitations of write endurance by arranging data access so that write operations can be distributed evenly across all the storage cells. During our intensive research on fault-tolerant NVM design, we found that ECC cannot effectively tolerate hard errors from limited write endurance and process imperfection. Therefore, we devised a novel Point and Discard (PAD) architecture in in [ 2012] as a hard-error-tolerant architecture for ReRAM-based Last Level Caches. PAD improves the lifetime of ReRAM caches by 1.6X-440X under different process variations without performance overhead in the system's early life. We have investigated the applicability of NVM for persistent memory design [Zhao 2013]. New byte addressable NVM enables fast persistent memory that allows in-memory persistent data objects to be updated with much higher throughput. Despite the significant improvement, the performance of these designs is only 50% of the native system with no persistence support, due to the logging or copy-on-write mechanisms used to update the persistent memory. A challenge in this approach is therefore how to efficiently enable atomic, consistent, and durable updates to ensure data persistence that survives application and/or system failures. We have designed a persistent memory system, called Klin, that can provide performance as close as that of the native system. The Klin design adopts a non-volatile cache and a non-volatile main memory for constructing a multi-versioned durable memory system, enabling atomic updates without logging or copy-on-write. Our evaluation shows that the proposed Kiln mechanism can achieve up to 2X of performance improvement to NVRAM-based persistent memory employing write-ahead logging. In addition, our design has numerous practical advantages: a simple and intuitive abstract interface, microarchitecture-level optimizations, fast recovery from failures, and no redundant writes to slow non-volatile storage media. The work was published in MICRO 2013 and received Best Paper Honorable Mentioned Award.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Roofline Analysis in the Intel® Advisor to Deliver Optimized Performance for applications on Intel® Xeon Phi™ Processor

In this session we show, in two case studies, how the roofline feature of Intel Advisor has been utilized to optimize the performance of kernels of the XGC1 and PICSAR codes in preparation for Intel Knights Landing architecture. The impact of the implemented optimizations and the benefits of using the automatic roofline feature of Intel Advisor to study performance of large applications will be presented. This demonstrates an effective optimization strategy that has enabled these science applications to achieve up to 4.6 times speed-up and prepare for future exascale architectures. # Goal/Relevance of Session The roofline model [1,2] is a powerful tool for analyzing the performance of applications with respect to the theoretical peak achievable on a given computer architecture. It allows one to graphically represent the performance of an application in terms of operational intensity, i.e. the ratio of flops performed and bytes moved from memory in order to guide optimization efforts. Given the scale and complexity of modern science applications, it can often be a tedious task for the user to perform the analysis on the level of functions or loops to identify where performance gains can be made. With new Intel tools, it is now possible to automate this task, as well as base the estimates of peak performance on measurements rather than vendor specifications. The goal of this session is to demonstrate how the roofline feature of Intel Advisor can be used to balance memory vs. computation related optimization efforts and effectively identify performance bottlenecks. A series of typical optimization techniques: cache blocking, structure refactoring, data alignment, and vectorization illustrated by the kernel cases will be addressed. # Description of the codes ## XGC1 The XGC1 code [3] is a magnetic fusion Particle-In-Cell code that uses an unstructured mesh for its Poisson solver that allows it to accurately resolve the edge plasma of a magnetic fusion device. After recent optimizations to its collision kernel [4], most of the computing time is spent in the electron push (pushe) kernel, where these optimization efforts have been focused. The kernel code scaled well with MPI+OpenMP but had almost no automatic compiler vectorization, in part due to indirect memory addresses and in part due to low trip counts of low-level loops that would be candidates for vectorization. Particle blocking and sorting have been implemented to increase trip counts of low-level loops and improve memory locality, and OpenMP directives have been added to vectorize compute-intensive loops that were identified by Advisor. The optimizations have improved the performance of the pushe kernel 2x on Haswell processors and 1.7x on KNL. The KNL node-for-node performance has been brought to within 30% of a NERSC Cori phase I Haswell node and we expect to bridge this gap by reducing the memory footprint of compute intensive routines to improve cache reuse. ## PICSAR is a Fortran/Python high-performance Particle-In-Cell library targeting at MIC architectures first designed to be coupled with the PIC code WARP for the simulation of laser-matter interaction and particle accelerators. PICSAR also contains a FORTRAN stand-alone kernel for performance studies and benchmarks. A MPI domain decomposition is used between NUMA domains and a tile decomposition (cache-blocking) handled by OpenMP has been added for shared-memory parallelism and better cache management. The so-called current deposition and field gathering steps that compose the PIC time loop constitute major hotspots that have been rewritten to enable more efficient vectorization. Particle communications between tiles and MPI domain has been merged and parallelized. All considered, these improvements provide speedups of 3.1 for order 1 and 4.6 for order 3 interpolation shape factors on KNL configured in SNC4 quadrant flat mode. Performance is similar between a node of cori phase 1 and KNL at order 1 and better on KNL by a factor 1.6 at order 3 with the considered test case (homogeneous thermal plasma).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Memory controller that forces prefetches in response to a present row address change timing constraint

An apparatus having a memory controller is described. The memory controller includes prefetch circuitry to prefetch, from a memory, data having a same row address in response to the memory controller's servicing of its request stream being stalled because of a timing constraint that prevents a change in row address. The memory controller also includes a cache to cache the prefetched data. The memory controller also includes circuitry to compare addresses of read requests in the memory controller's request stream against respective addresses of the prefetched data in the cache and to service those of the requests in the memory controller's request stream having a matching address with corresponding ones of the prefetched data in the cache.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

$\mathrm{PPT}$-Multicore: performance prediction of Open$\mathrm{MP}$ applications using reuse profiles and analytical modeling

In this report we present PPT-Multicore, an analytical model embedded in the Performance Prediction Toolkit (PPT) to predict parallel applications’ performance running on a multicore processor. PPT-Multicore builds upon our previous work towards a multicore cache model. We extract LLVM basic block labeled memory trace using an architecture-independent LLVM-based instrumentation tool only once in an application’s lifetime. The model uses the memory trace and other parameters from an instrumented sequentially executed binary. We use probabilistic and computationally efficient reuse profiles to predict the cache hit rates and runtimes of OpenMP programs’ parallel sections. We model Intel’s Broadwell, Haswell, and AMD’s Zen2 architectures and validate our framework using different applications from PolyBench and PARSEC benchmark suites. The results show that PPT-Multicore can predict cache hit rates with an overall average error rate of 1.23% while predicting the runtime with an error rate of 9.08%.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

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.↗
DOE OSTI2017

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↗
DOE OSTI2018

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↗
DOE OSTI2019

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↗
DOE OSTI2023

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↗
DOE OSTI2023

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↗
DOE OSTI2018

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↗
DOE OSTI2026

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↗
DOE OSTI2018

Multiple-core computer processor for reverse time migration

A multi-core computer processor including a plurality of processor cores interconnected in a Network-on-Chip (NoC) architecture, a plurality of caches, each of the plurality of caches being associated with one and only one of the plurality of processor cores, and a plurality of memories, each of the plurality of memories being associated with a different set of at least one of the plurality of processor cores and each of the plurality of memories being configured to be visible in a global memory address space such that the plurality of memories are visible to two or more of the plurality of processor cores, wherein at least one of a number of the processor cores, a size of each of the plurality of caches, or a size of each of the plurality of memories is configured for performing a reverse-time-migration (RTM) computation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Resolution system that provides access to remote data locally

A system and method that provides access to remote data locally by identifying and downloading linked content from a plurality of remote sources through a communication network; decomposing the downloaded content into local content elements by intercepting links between the content elements and mapping the intercepted links to redirected links that locate the same download content in a local cache or a local server proxy; identifying remote devices requesting access to the local cache or the local server proxy and the device's functionality through a tracking code stored on the remote devices and a probabilistic tracking; and providing access to the local cache or the local server proxy in response to authorizations associated with the identified devices and a user.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Method and apparatus for maintaining data coherence in a non-uniform compute device

A data processing apparatus includes one or more host processors with first processing units, one or more caches with second processing unit, a non-cache memory having a third processing unit and a reorder buffer operable to maintain data order during execution of a program of instructions. An instruction scheduler routes instructions to the processing units. Data coherence is maintained by control logic that blocks access to data locations in use by a selected processing unit other than the selected processing unit until data associated with the data locations are released from the reorder buffer. Data stored in the cache is written to the memory if it is already in a modified state, otherwise the state is set to the modified state. A memory controller may be used to restrict access to memory locations to be operated on.

97 MATHEMATICS AND COMPUTING↗