CoNDA: Efficient cache coherence support for near-data accelerators
Specialized on-chip accelerators are widely used to improve the energy efficiency of
computing systems. Recent advances in memory technology have enabled near-data …
computing systems. Recent advances in memory technology have enabled near-data …
Crow: A low-cost substrate for improving dram performance, energy efficiency, and reliability
DRAM has been the dominant technology for architecting main memory for decades. Recent
trends in multi-core system design and large-dataset applications have amplified the role of …
trends in multi-core system design and large-dataset applications have amplified the role of …
Non-relational databases on FPGAs: Survey, design decisions, challenges
Non-relational database systems (NRDS) such as graph and key-value have gained
attention in various trending business and analytical application domains. However, while …
attention in various trending business and analytical application domains. However, while …
Demystifying memory access patterns of FPGA-based graph processing accelerators
Recent advances in reprogrammable hardware (eg, FPGAs) and memory technology (eg,
DDR4, HBM) promise to solve performance problems inherent to graph processing like …
DDR4, HBM) promise to solve performance problems inherent to graph processing like …
Automatic sublining for efficient sparse memory accesses
Sparse memory accesses, which are scattered accesses to single elements of a large data
structure, are a challenge for current processor architectures. Their lack of spatial and …
structure, are a challenge for current processor architectures. Their lack of spatial and …
Data-Centric and Data-Aware Frameworks for Fundamentally Efficient Data Handling in Modern Computing Systems
N Ha**azar - ar** for emerging memory devices
Recent advancements in 3D-stacked DRAM such as hybrid memory cube (HMC) and high-
bandwidth memory (HBM) promise higher bandwidth and lower power consumption …
bandwidth memory (HBM) promise higher bandwidth and lower power consumption …
FPGA-based Query Acceleration for Non-relational Databases
JC Dann - 2024 - archiv.ub.uni-heidelberg.de
Database management systems are an integral part of today's everyday life. Trends like
smart applications, the internet of things, and business and social networks require …
smart applications, the internet of things, and business and social networks require …
Walter: wide I/O scaling of number of memory controllers versus frequency and voltage
MD Marino - IEEE Access, 2020 - ieeexplore.ieee.org
Computational application demands do push the scaling of the number of cores, which
themselves further increase the demand for more bandwidth. The use of larger rank widths …
themselves further increase the demand for more bandwidth. The use of larger rank widths …
McSimA+ 시뮬레이터를 사용한 Vision Transformer 추론 과정의 레이어 별 Memory Bottleneck 분석
황인성， 장지훈， 신진， 김현 - 대한전자공학회 학술대회, 2023 - dbpia.co.kr
As deep learning models continue to grow in scale, the number of parameters in these
models has increased, causing a significant memory bottleneck in conventional von …
models has increased, causing a significant memory bottleneck in conventional von …