Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 19 of 19 for “"Memory-level parallelism"”.

  1. Software prefetching for memory-level parallelism

    … latency. Existing techniques to tolerate long memory latencies include data prefetching, out-of-order instruction execution, and multithreading. However, because of limited buffers, purely hardware-based mechanisms are not capable of exploiting sufficient memory-level parallelism (MLP) to …

    texas Repository record for Software prefetching for memory-level parallelism (opens in a new tab)

  2. Efficient memory-level parallelism extraction with decoupled strands

    … processors that exploits intra-thread memory-level parallelism (MLP) to improve performance efficiency on highly threaded workloads. Outrider enables a single thread of execution to be presented to the architecture as multiple decoupled instruction streams, consisting of either memory …

    uiuc Repository record for Efficient memory-level parallelism extraction with decoupled strands (opens in a new tab)

  3. Exploiting Memory-Level Parallelism to Mitigate the Impact of Critical Loads on Performance and Energy Efficiency

    … behind useful work from other threads (thread-level parallelism – TLP) by context switching after encountering critical loads and paying the context-switching cost. The context switch can be triggered after a load stalls the ROB (reactive approach), but this limits gains from TLP. Instead, we …

    gatech Repository record for Exploiting Memory-Level Parallelism to Mitigate the Impact of Critical Loads on Performance and Energy Efficiency (opens in a new tab)

  4. Group Operations for Concurrent Binary Search Trees

    … that improves throughput by overlapping memory access latencies for independent traversals. Constructing code that maximizes this throughput benefit by fully leveraging the instruction- and memory-level parallelism available on modern hardware is challenging because of overhead, …

    portland-state Repository record for Group Operations for Concurrent Binary Search Trees (opens in a new tab)

  5. Panther : a memory-centric accelerator for webscale graph processing

    … limited by irregular, data-dependent memory accesses. These access patterns do not align well with cache hierarchies and leave the compute resources of traditional CPUs and GPUs underutilized. Conventional architectures struggle to provide the necessary concurrency to mask these …

    texas Repository record for Panther : a memory-centric accelerator for webscale graph processing (opens in a new tab)

  6. Toward Adaptive Modular Computing Systems Through More Expressive and Elastic Hardware and Software Interfaces

    … productive specialization, allowing processors, memory technologies, network interfaces, and software stacks to evolve on independent trajectories without requiring tight coordination. The cost of this independence, however, is that each component is designed and provisioned based on static …

    gatech Repository record for Toward Adaptive Modular Computing Systems Through More Expressive and Elastic Hardware and Software Interfaces (opens in a new tab)

  7. Criticality driven execution

    … through accurate branch prediction and reducing memory access latencies using data prefetchers and a cache hierarchy. Despite advancements in prediction algorithms and data prefetching techniques, the remaining branch mispredictions and cache misses still present major bottlenecks in many …

    texas Repository record for Criticality driven execution (opens in a new tab)

  8. A Manycore Architecture for Scalability, Programmability, and Compute Density

    … as VLSI resource organization, on-chip networks, memory hierarchy, memory-level parallelism, address mapping and translation, synchronization, and thread management. Its flexibility has been demonstrated with a parallel benchmark suite representing a broad range of computational and communication …

    washington Repository record for A Manycore Architecture for Scalability, Programmability, and Compute Density (opens in a new tab)

  9. Inter-core locality aware memory access scheduling

    … thousands of parallel threads and achieve high Memory Level Parallelism (MLP). To support high MLP, a structure called a Miss-Status Holding Register (MSHR) handles multiple in-flight miss requests. When multiple cores send requests to the same cache line, the requests are merged into one last …

    ubc Repository record for Inter-core locality aware memory access scheduling (opens in a new tab) Download PDF for Inter-core locality aware memory access scheduling (opens in a new tab)

  10. Criticality-Aware Memory Systems

    Research on computer memory systems has been of increasing importance over the last decade, as they have become a significant bottleneck for application performance. While newer memory systems offer increased memory level parallelism, they cannot be used blindly due to contention for shared …

    cornell Repository record for Criticality-Aware Memory Systems (opens in a new tab)

  11. Performance and energy efficiency via an adaptive MorphCore architecture

    The level of Thread-Level Parallelism (TLP), Instruction-Level Parallelism (ILP), and Memory-Level Parallelism (MLP) varies across programs and across program phases. Hence, every program requires different underlying core microarchitecture resources for high performance and/or energy efficiency. …

    texas Repository record for Performance and energy efficiency via an adaptive MorphCore architecture (opens in a new tab)

  12. Cross-stack Improvement on Memory Efficiency

    Modern data centers increasingly host memory-intensive applications that demand vast memory capacity. However, memory performance has not scaled at the same pace as compute performance, and indiscriminately expanding memory resources incurs substantial infrastructure and maintenance costs, …

    vt Repository record for Cross-stack Improvement on Memory Efficiency (opens in a new tab)

  13. Modeling performance and power for energy-efficient GPGPU computing

    … relationship that exists between the thread-level parallelism and memory-level parallelism for an application on a given many-core architecture. Two metrics are proposed: memory warp parallelism (MWP), which refers to the number of overlapping memory accesses per core, and computation warp …

    gatech Repository record for Modeling performance and power for energy-efficient GPGPU computing (opens in a new tab)

  14. Adaptive caching for high-performance memory systems

    … system performance has been the access to main memory, which is typically two orders of magnitude slower than the processor. To bridge this gap, modern processors already devote more than half of the on-chip transistors to the last-level cache. However, traditional cache designs – developed for …

    texas Repository record for Adaptive caching for high-performance memory systems (opens in a new tab)

  15. Time-Division-Multiplexing Based Hybrid-Switched NoC for Heterogeneous Multicore Systems

    … the dissertation focuses on maintaining global memory access order using the proposed hybrid-switched NoC, allowing memory operations to perform in parallel while a stronger memory consistency model can still be satisfied. The memory consistency model specifies the order how reads and writes …

    umn Repository record for Time-Division-Multiplexing Based Hybrid-Switched NoC for Heterogeneous Multicore Systems (opens in a new tab)

  16. Performance Exploration of the Hybrid Memory Cube

    The Hybrid Memory Cube (HMC) is an emerging main memory technology that leverages advances in 3D fabrication techniques to create a memory device with several DRAM dies stacked on top of a CMOS logic layer. The logic layer at the base of each stack contains several DRAM memory controllers that …

    maryland Repository record for Performance Exploration of the Hybrid Memory Cube (opens in a new tab)

  17. Exploiting on-chip memory concurrency in 3d manycore architectures

    … technology trends point to large increases in memory-level concurrency. This in turn affects the design of the multi-core interconnect and organization of the memory hierarchy. The work addresses the need for re-optimization in the presence of this increase in concurrency of the memory system. …

    gatech Repository record for Exploiting on-chip memory concurrency in 3d manycore architectures (opens in a new tab)

  18. Memory system design and optimizations for data intensive computing

    … exhibit tremendous datasets and irregular memory access patterns. Research has shown that these memory-bound applications are unable to effectively leverage the principles of data locality and regular memory accesses within traditional cache-based memory systems to mitigate the …

    ttu Repository record for Memory system design and optimizations for data intensive computing (opens in a new tab)

  19. Towards achieving predictable memory performance on multi-core based mixed criticality embedded systems

    … tasks. The shared resources such as Memory subsystem (caches and DRAM) and, Bus make this a challenging effort. This work focuses on the shared Memory subsystem. First, we studied Commercial-Off-The-Shelf (COTS) DRAM controllers to an extend to demonstrate the interference due to …

    ku Repository record for Towards achieving predictable memory performance on multi-core based mixed criticality embedded systems (opens in a new tab)