Výsledky vyhľadávania - single instruction multiple data parallel computation model

A GPU-based numerical manifold method for modeling the formation of the excavation damaged zone in deep rock tunnels Autor Liu, Quanshen, Xu, Xiangyu, Wu, Zhijun

“…In this study, combined with the zero–thickness cohesive element (ZE) model and explicit integration method, a parallelization technique based on graphics processing units (GPU…”
Získať plný text

Splitwise: Efficient Generative LLM Inference Using Phase Splitting Autor Patel, Pratyush, Choukse, Esha, Zhang, Chaojie, Shah, Aashaka, Goiri, Inigo, Maleki, Saeed, Bianchini, Ricardo

“…Generative large language model (LLM) applications are growing rapidly, leading to large-scale deployments of expensive and power-hungry GPUs…”
Získať plný text

Max-PIM: Fast and Efficient Max/Min Searching in DRAM Autor Zhang, Fan, Angizi, Shaahin, Fan, Deliang

“…Recently, in-DRAM computing is becoming one promising technique to address the notorious 'memory-wall' issue for big data processing…”
Získať plný text

A flexible algorithm for calculating pair interactions on SIMD architectures Autor Páll, Szilárd, Hess, Berk

“… In order to reach high performance on modern CPU and accelerator architectures, single-instruction multiple-data (SIMD…”
Získať plný text

PISA: Efficient Precision-Slice Framework for LLMs with Adaptive Numerical Type Autor Yang, Ning, Wang, Zongwu, Sun, Qingxiao, Lu, Liqiang, Liu, Fangxin

“…Large language models (LLMs) have transformed numerous AI applications, with on-device deployment becoming increasingly important for reducing cloud computing costs and protecting user privacy…”
Získať plný text

McPAL: Scaling Unstructured Sparse Inference with Multi-Chiplet HBM-PIM Architecture for LLMs Autor Liu, Shiwei, Huang, Zhirui, Yu, Jiangnan, Liu, Qi, Chen, Chixiao

“…Large language models (LLMs) have gained significant attention recently. However, executing LLM is memory-bound due to the extensive memory accesses. Process-in-memory (PIM…”
Získať plný text

KVO-LLM: Boosting Long-Context Generation Throughput for Batched LLM Inference Autor Li, Zhenyu, Lyu, Dongxu, Wang, Gang, Chen, Yuzhou, Chen, Liyan, Li, Wenjie, Jiang, Jianfei, Sun, Yanan, He, Guanghui

“…With the widespread deployment of long-context large language models (LLMs), efficient and high-quality generation is becoming increasingly important…”
Získať plný text

A Novel Wavefront-Based High Parallel Solution for HEVC Encoding Autor Chen, Keji, Sun, Jun, Duan, Yizhou, Guo, Zongming

“… On data level, optimal single-instruction-multiple-data algorithms are designed for the enhanced coding tools, i.e…”
Získať plný text

MIMDRAM: An End-to-End Processing-Using-DRAM System for High-Throughput, Energy-Efficient and Programmer-Transparent Multiple-Instruction Multiple-Data Computing Autor Oliveira, Geraldo F., Olgun, Ataberk, Yaglikci, Abdullah Giray, Bostanci, F. Nisa, Gomez-Luna, Juan, Ghose, Saugata, Mutlu, Onur

“…, 16,384-262,144-bit-wide) data-parallel operations, in a single-instruction multiple-data (SIMD) fashion. However, DRAM rows' large and rigid granularity limit the effectiveness and applicability of PUD in three ways…”
Získať plný text

PHCG: Optimizing Simulink Code Generation for Embedded System With SIMD Instructions Autor Su, Zhuo, Wang, Dongyan, Yu, Zehong, Yang, Yixiao, Jiang, Yu, Wang, Rui, Chang, Wanli, Li, Wen, Cui, Aiguo, Sun, Jiaguang

“… In this article, we propose PHCG, an optimized code generator for the Simulink model with single-instruction-multiple-data (SIMD…”
Získať plný text

PairGraph: An Efficient Search-space-aware Accelerator for High-performance Concurrent Pairwise Queries Autor Fu, Yutao, Long, Zhongtian, Zhang, Yu, He, Zirui, Zhao, Jin, Niu, Qiyuan, Wang, Zixiao, Jin, Hai

“…Pairwise queries have been widely used in many applications. Although several approaches have been recently proposed to accelerate a single query, they still…”
Získať plný text

Pipirima: Predicting Patterns in Sparsity to Accelerate Matrix Algebra Autor Bakhtiar, Ubaid, Joo, Donghyeon, Asgari, Bahar

“…While sparsity, a feature of data in many applications, provides optimization opportunities such as reducing unnecessary computations, data transfers, and storage, it causes several challenges, too…”
Získať plný text

PacQ: A SIMT Microarchitecture for Efficient Dataflow in Hyper-asymmetric GEMMs Autor Yin, Ruokai, Li, Yuhang, Panda, Priyadarshini

“… During deployment on single-instruction-multiple-threads (SIMT) architectures, weights are stored in low-precision integer (INT…”
Získať plný text

PIMGCN: A ReRAM-Based PIM Design for Graph Convolutional Network Acceleration Autor Yang, Tao, Li, Dongyue, Han, Yibo, Zhao, Yilong, Liu, Fangxin, Liang, Xiaoyao, He, Zhezhi, Jiang, Li

“…Graph Convolutional Network (GCN) is a promising but computing- and memory-intensive learning model…”
Získať plný text

A parallel algorithm for generating bicompatible elimination orderings of proper interval graphs Autor Panda, B.S., Das, Sajal K.

“… (Single Instruction Stream Multiple Data Stream Concurrent Read Concurrent Write Parallel Random Access Machine…”
Získať plný text

EDGE: Event-Driven GPU Execution Autor Hetherington, Tayler Hicklin, Lubeznov, Maria, Shah, Deval, Aamodt, Tor M.

“… supporting latency-sensitive streaming tasks. This paper proposes an event-driven GPU execution model, EDGE, that enables non-CPU devices to directly launch preconfigured tasks on a GPU without CPU interaction…”
Získať plný text

Occamy: Memory-efficient GPU Compiler for DNN Inference Autor Lee, Jaeho, Jeong, Shinnung, Song, Seungbin, Kim, Kunwoo, Choi, Heelim, Kim, Youngsok, Kim, Hanjun

“…This work proposes Occamy, a new memory-efficient DNN compiler that reduces the memory usage of a DNN model without affecting its accuracy…”
Získať plný text

Soter: Analytical Tensor-Architecture Modeling and Automatic Tensor Program Tuning for Spatial Accelerators Autor Wang, Fuyu, Shen, Minghua, Ding, Yufei, Xiao, Nong

“…Spatial accelerator is a specialized hardware to provide noticeable performance speedup for tensor computations…”
Získať plný text

Offloaded MPI message matching: an optimistic approach Autor Garcia, Jeronimo S., Di Girolamo, Salvatore, Kosta, Sokol, Olmos, J.J. Vegas, Nudelman, Rami, Hoefler, Torsten, Bloch, Gil

“… In this work, we propose a bin-based MPI message approach, Optimistic Tag Matching, explicitly designed for the lightweight, highly parallel architectures typical of on-path SmartNICs…”
Získať plný text

Accelerating Geostatistical Modeling and Prediction With Mixed-Precision Computations: A High-Productivity Approach With PaRSEC Autor Abdulah, Sameh, Cao, Qinglei, Pei, Yu, Bosilca, George, Dongarra, Jack, Genton, Marc G., Keyes, David E., Ltaief, Hatem, Sun, Ying

“… models and optimization of parameters. Spatial data are assumed to possess properties of stationarity or non-stationarity via a kernel fitted to a covariance matrix…”
Získať plný text