An official website of the United States government
In this paper, we propose GraphiDe, a novel DRAM-based processing-in-memory (PIM) accelerator for graph processing. It transforms current DRAM architecture to massively parallel computational units exploiting the high internal bandwidth of the modern memory chips to accelerate various graph processing applications. GraphiDe can be leveraged to greatly reduce energy consumption and latency dealing with underlying adjacency matrix computations by eliminating unnecessary off-chip accesses. The extensive circuit-architecture simulations over three social network data-sets indicate that GraphiDe achieves on average 3.1x energy-efficiency improvement and 4.2x speed-up over the recent DRAM based PIM platform. It achieves ~59x higher energy-efficiency and 83x speed-up over GPU-based acceleration methods.
more »
« less
Scalable Post-Processing of Large-Scale Numerical Simulations of Turbulent Fluid Flows
Military, space, and high-speed civilian applications will continue contributing to the renewed interest in compressible, high-speed turbulent boundary layers. To further complicate matters, these flows present complex computational challenges ranging from the pre-processing to the execution and subsequent post-processing of large-scale numerical simulations. Exploring more complex geometries at higher Reynolds numbers will demand scalable post-processing. Modern times have brought application developers and scientists the advent of increasingly more diversified and heterogeneous computing hardware, which significantly complicates the development of performance-portable applications. To address these challenges, we propose Aquila, a distributed, out-of-core, performance-portable post-processing library for large-scale simulations. It is designed to alleviate the burden of domain experts writing applications targeted at heterogeneous, high-performance computers with strong scaling performance. We provide two implementations, in C++ and Python; and demonstrate their strong scaling performance and ability to reach 60% of peak memory bandwidth and 98% of the peak filesystem bandwidth while operating out of core. We also present our approach to optimizing two-point correlations by exploiting symmetry in the Fourier space. A key distinction in the proposed design is the inclusion of an out-of-core data pre-fetcher to give the illusion of in-memory availability of files yielding up to 46% improvement in program runtime. Furthermore, we demonstrate a parallel efficiency greater than 70% for highly threaded workloads.
more »
« less
- Award ID(s):
- 1849243
- PAR ID:
- 10322171
- Date Published:
- Journal Name:
- Symmetry
- Volume:
- 14
- Issue:
- 4
- ISSN:
- 2073-8994
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)As FPGA-based accelerators become ubiquitous and more powerful, the demand for integration with High-Performance Memory (HPM) grows. Although HPMs offer a much greater bandwidth than standard DDR4 DRAM, they introduce new design challenges such as increased latency and higher bandwidth mismatch between memory and FPGA cores. This paper presents a scalable architecture for convolutional neural network accelerators conceived specifically to address these challenges and make full use of the memory's high bandwidth. The accelerator, which was designed using high-level synthesis, is highly configurable. The intrinsic parallelism of its architecture allows near-perfect scaling up to saturating the available memory bandwidth.more » « less
-
The memory footprint of modern applications like large language models (LLMs) far exceeds the memory capacity of accelerators they run on and often spills over to host memory. As model sizes continue to grow, DRAM-based memory is no longer sufficient to contain these models, resulting in further spill-over to storage and necessitating the use of technologies like Intel Optane and CXL-enabled memory expansion. While such technologies provide more capacity, their higher latency and lower bandwidth has given rise to heterogeneous memory configurations that attempt to strike a balance between capacity and performance. This paper evaluates the impact of such memory configurations on a GPU running out-of-core LLMs. Starting with basic host/device bandwidth measurements using an Optane and Nvidia A100 equipped NUMA system, we present a comprehensive performance analysis of serving OPT-30B and OPT-175B models using FlexGen, a state-of-the-art serving framework. Our characterization shows that FlexGen's weight placement algorithm is a key bottleneck limiting performance. Based on this observation, we evaluate two alternate weight placement strategies, one each optimizing for inference latency and throughput. When combined with model quantization, our strategies improve latency and throughput by 27% and 5x, respectively. These figures are within 9% and 6% of an all-DRAM system, demonstrating how careful data placement can effectively enable the substitution of DRAM with high-capacity but slower memory, improving overall system energy efficiency.more » « less
-
System on chips (SoCs) are now designed with their own artificial intelligence (AI) accelerator segment to accommodate the ever-increasing demand of deep learning (DL) applications. With powerful multiply and accumulate (MAC) engines for matrix multiplications, these accelerators show high computing performance. However, because of limited memory resources (i.e., bandwidth and capacity), they fail to achieve optimum system performance during large batch training and inference. In this work, we propose a memory system with high on-chip capacity and bandwidth to shift the gear of AI accelerators from memory-bound to achieving system-level peak performance. We develop the memory system with design technology co-optimization (DTCO)-enabled customized spin-orbit torque (SOT)-MRAM as large on-chip memory through system technology co-optimization (STCO) and detailed characterization of the DL workloads. Our workload-aware memory system achieves 8× energy and 9× latency improvement on computer vision (CV) benchmarks in training and 8× energy and 4.5× latency improvement on natural language processing (NLP) benchmarks in training while consuming only around 50% of SRAM area at iso-capacity.more » « less
-
As scaling of conventional memory devices has stalled, many high-end computing systems have begun to incorporate alternative memory technologies to meet performance goals. Since these technologies present distinct advantages and tradeoffs compared to conventional DDR* SDRAM, such as higher bandwidth with lower capacity or vice versa, they are typically packaged alongside conventional SDRAM in a heterogeneous memory architecture. To utilize the different types of memory efficiently, new data management strategies are needed to match application usage to the best available memory technology. However, current proposals for managing heterogeneous memories are limited, because they either (1) do not consider high-level application behavior when assigning data to different types of memory or (2) require separate program execution (with a representative input) to collect information about how the application uses memory resources. This work presents a new data management toolset to address the limitations of existing approaches for managing complex memories. It extends the application runtime layer with automated monitoring and management routines that assign application data to the best tier of memory based on previous usage, without any need for source code modification or a separate profiling run. It evaluates this approach on a state-of-the-art server platform with both conventional DDR4 SDRAM and non-volatile Intel Optane DC memory, using both memory-intensive high-performance computing (HPC) applications as well as standard benchmarks. Overall, the results show that this approach improves program performance significantly compared to a standard unguided approach across a variety of workloads and system configurations. The HPC applications exhibit the largest benefits, with speedups ranging from 1.4× to 7× in the best cases. Additionally, we show that this approach achieves similar performance as a comparable offline profiling-based approach after a short startup period, without requiring separate program execution or offline analysis steps.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/sym14040823
