Suchergebnisse - Theory of computation → Shared memory algorithms

Verifying Lock-Free Search Structure Templates (Artifact) von Patel, Nisarg, Shasha, Dennis, Wies, Thomas

“… We present and verify template algorithms for lock-free concurrent search structures that cover a broad range of existing implementations based on lists and skiplists …”
Volltext

pSyncPIM: Partially Synchronous Execution of Sparse Matrix Operations for All-Bank PIM Architectures von Baek, Daehyeon, Hwang, Soojin, Huh, Jaehyuk

“… Recent commercial incarnations of processing-in-memory (PIM) maintain the standard DRAM interface and employ the all-bank mode execution to maximize bank-level memory bandwidth …”
Volltext

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

“… In this work, for the first time, we propose a novel 'Min/Max-in-memory' algorithm based on iterative XNOR bit-wise comparison, which supports parallel inmemory searching for minimum and maximum …”
Volltext

SplitSync: Bank Group-Level Split-Synchronization for High-Performance DRAM PIM von Yoon, Byungkuk, Han, Sanghyeok, Park, Gyeonghwan, Kim, Jae-Joon

“… Processing in Memory (PIM) architectures enhance memory bandwidth by utilizing bank-level parallelism, typically implemented with a SIMD structure where all banks operate simultaneously under a single command …”
Volltext

PIMDup: An Optimized Deduplication Design on a Real Processing-in-Memory System von Yeh, Chun-Le, Chen, Liang-Chi, Ho, Chien-Chung, Chang, Yu-Ming, Chang, Da-Wei

“… To overcome these challenges, we explore UPMEM's DPU, a processing-in-memory (PIM) technology that reduces data movement by performing computations directly within memory …”
Volltext

UPVSS: Jointly Managing Vector Similarity Search with Near-Memory Processing Systems von Liu, Chun-Chien, Wu, Chun-Feng, Jin, Yunho

“… to performance limitations in conventional von Neumann architecture with shared memory buses. This data movement bottleneck restricts the efficiency and scalability of vector similarity search due to insufficient memory bandwidth …”
Volltext

AttenPIM: Accelerating LLM Attention with Dual-mode GEMV in Processing-in-Memory von Chen, Liyan, Lyu, Dongxu, Li, Zhenyu, Jiang, Jianfei, Wang, Qin, Mao, Zhigang, Jing, Naifeng

“… While recent heterogeneous architectures attempt to address the memory-bound bottleneck from attention computations by processing-in-memory (PIM …”
Volltext

BLOwing Trees to the Ground: Layout Optimization of Decision Trees on Racetrack Memory von Hakert, Christian, Khan, Asif Ali, Chen, Kuan-Hsun, Hameed, Fazal, Castrillon, Jeronimo, Chen, Jian-Jia

“… Modern distributed low power systems tend to integrate machine learning algorithms, which are directly executed on the distributed devices (on the edge …”
Volltext

SDM: Sharing-Enabled Disaggregated Memory System with Cache Coherent Compute Express Link von Lee, Hyokeun, Choi, Kwanseok, Lee, Hvuk-Jae, Sim, Jaewoong

“… Disaggregated memory has been gaining significant traction as a promising solution for scaling memory capacity and better utilizing memory resources in data centers …”
Volltext

Hardware Support for Durable Atomic Instructions for Persistent Parallel Programming von Hadi, Khan Shaikhul, Ul Mustafa, Naveed, Heinrich, Mark, Solihin, Yan

“… Persistent memory is emerging as an attractive main memory fabric capable of hosting persistent data …”
Volltext

Optimizing Persistent Memory Transactions von Zardoshti, Pantea, Zhou, Tingzhe, Liu, Yujie, Spear, Michael

“… Byte-addressable, non-volatile, random access memory (NVM) has the potential to dramatically accelerate the performance of storage-intensive workloads …”
Volltext

Separating Mechanism from Policy in STM von Sheng, Yaodong, Hassan, Ahmed, Spear, Michael

“… On one hand, Software Transactional Memory (STM) is easy, because it allows programmers to simply mark regions of sequential code as requiring atomicity, and then the compiler ensures that no races manifest …”
Volltext

DARIC: A Data Reuse-Friendly CGRA for Parallel Data Access via Elastic FIFOs von Liu, Dajiang, Mou, Di, Zhu, Rong, Zhuang, Yan, Shang, Jiaxing, Zhong, Jiang, Yin, Shouyi

“… For parallel data accesses, uniform memory partitioning is usually introduced to CGRA for better pipelining performance …”
Volltext

Asynchronous Distributed-Memory Parallel Algorithms for Influence Maximization von Singhal, Shubhendra Pal, Hati, Souvadra, Young, Jeffrey, Sarkar, Vivek, Hayashi, Akihiro, Vuduc, Richard

“… We propose distributed-memory parallel algorithms for the two main kernels of a state-of-the-art implementation of one IM algorithm, influence maximization via martingales (IMM …”
Volltext

Ultrafast CPU/GPU Kernels for Density Accumulation in Placement von Guo, Zizheng, Mai, Jing, Lin, Yibo

“… Density accumulation is a widely-used primitive operation in physical design, especially for placement. Iterative invocation in the optimization flow makes it …”
Volltext

Scalable task parallelism for NUMA: A uniform abstraction for coordinated scheduling and memory management von Drebes, Andi, Pop, Antoniu, Heydemann, Karine, Cohen, Albert, Drach, Nathalie

“… Dynamic task-parallel programming models are popular on shared-memory systems, promising enhanced scalability, load balancing and locality …”
Volltext

Unfair Scheduling Patterns in NUMA Architectures von Ben-David, Naama, Scully, Ziv, Blelloch, Guy E.

“… Lock-free algorithms are typically designed and analyzed with adversarial scheduling in mind …”
Volltext

CAF: Core to core Communication Acceleration Framework von Yipeng Wang, Ren Wang, Herdrich, Andrew, Tsai, James, Solihin, Yan

“… The traditional way cores communicate is by using shared memory space between them. However, shared memory communication fundamentally involves coherence invalidations …”
Volltext

Enumeration of Billions of Maximal Bicliques in Bipartite Graphs without Using GPUs von Pan, Zhe, He, Shuibing, Li, Xu, Zhang, Xuechen, Yin, Yanlong, Wang, Rui, Shou, Lidan, Song, Mingli, Sun, Xian-He, Chen, Gang

“… To overcome this limitation, we propose an AdaMBE algorithm. First, we redesign its core operations using local neighborhood information derived from computational subgraphs to minimize redundant memory accesses …”
Volltext

Parallel Top-K Algorithms on GPU: A Comprehensive Study and New Methods von Zhang, Jingrong, Naruse, Akira, Li, Xipeng, Wang, Yong

“… As data volume grows rapidly, high-performance parallel top-K algorithms become critical …”
Volltext