Partitioning and Data Mapping in Reconfigurable Cache and Scratchpad Memory--Based Architectures

2016 ◽  
Vol 22 (1) ◽  
pp. 1-25 ◽  
Author(s):  
Prasenjit Chakraborty ◽  
Preeti Ranjan Panda ◽  
Sandeep Sen
Keyword(s):  
Data Mapping ◽  
Scratchpad Memory ◽  
Reconfigurable Cache

scholarly journals Decision Tree-Based Adaptive Reconfigurable Cache Scheme

Algorithms ◽  
2021 ◽  
Vol 14 (6) ◽  
pp. 176
Author(s):  
Wei Zhu ◽  
Xiaoyang Zeng
Keyword(s):  
Decision Tree ◽  
Adaptive Algorithms ◽  
Memory Access ◽  
Access Time ◽  
Decision Tree Algorithm ◽  
Verilog Hdl ◽  
Tree Model ◽  
Cache Associativity ◽  
Cache Scheme ◽  
Reconfigurable Cache

Applications have different preferences for caches, sometimes even within the different running phases. Caches with fixed parameters may compromise the performance of a system. To solve this problem, we propose a real-time adaptive reconfigurable cache based on the decision tree algorithm, which can optimize the average memory access time of cache without modifying the cache coherent protocol. By monitoring the application running state, the cache associativity is periodically tuned to the optimal cache associativity, which is determined by the decision tree model. This paper implements the proposed decision tree-based adaptive reconfigurable cache in the GEM5 simulator and designs the key modules using Verilog HDL. The simulation results show that the proposed decision tree-based adaptive reconfigurable cache reduces the average memory access time compared with other adaptive algorithms.

scholarly journals Thread-Aware Mechanism to Enhance Inter-Node Load Balancing for Multithreaded Applications on NUMA Systems

2021 ◽  
Vol 11 (14) ◽  
pp. 6486
Author(s):  
Mei-Ling Chiang ◽  
Wei-Lun Su
Keyword(s):  
Load Balancing ◽  
Critical Path ◽  
Selection Procedure ◽  
Remote Memory ◽  
Data Mapping ◽  
Linux Kernel ◽  
Access Memory ◽  
Selection Policy ◽  
Benchmark Suite ◽  
Thread Mapping

NUMA multi-core systems divide system resources into several nodes. When an imbalance in the load between cores occurs, the kernel scheduler’s load balancing mechanism then migrates threads between cores or across NUMA nodes. Remote memory access is required for a thread to access memory on the previous node, which degrades performance. Threads to be migrated must be selected effectively and efficiently since the related operations run in the critical path of the kernel scheduler. This study focuses on improving inter-node load balancing for multithreaded applications. We propose a thread-aware selection policy that considers the distribution of threads on nodes for each thread group while migrating one thread for inter-node load balancing. The thread is selected for which its thread group has the least exclusive thread distribution, and thread members are distributed more evenly on nodes. This has less influence on data mapping and thread mapping for the thread group. We further devise several enhancements to eliminate superfluous evaluations for multithreaded processes, so the selection procedure is more efficient. The experimental results for the commonly used PARSEC 3.0 benchmark suite show that the modified Linux kernel with the proposed selection policy increases performance by 10.7% compared with the unmodified Linux kernel.

High Throughput Data Mapping for Coarse-Grained Reconfigurable Architectures

2011 ◽  
Vol 30 (11) ◽  
pp. 1599-1609 ◽  
Author(s):  
Yongjoo Kim ◽  
Jongeun Lee ◽  
A. Shrivastava ◽  
J. W. Yoon ◽  
Doosan Cho ◽  
...  
Keyword(s):  
High Throughput ◽  
Coarse Grained ◽  
Reconfigurable Architectures ◽  
Data Mapping ◽  
High Throughput Data

Modeling and Implementation of Inter-Campuses Equipment Optimized Management Based on WebGIS

2013 ◽  
Vol 680 ◽  
pp. 534-539
Author(s):  
Wei Feng Ma
Keyword(s):  
Information Management ◽  
Spatial Data ◽  
Large Scale ◽  
New Technology ◽  
Rapid Expansion ◽  
Mathematics Model ◽  
Data Mapping ◽  
Geographically Dispersed ◽  
New Research ◽  
Research Challenge

With the rapid expansion of the campus scale and the increasing of the geographically dispersed campus, how to adopt new theory, new method and new technology to realize the equipment optimized assignment and the information management is a new research challenge. It is the key to safeguard the national fund to use reasonably, and to speed up the development of education healthily. Through analyzing the domestic and foreign related research works, the paper proposed that it can take use of the spatial data expression and analysis with Geographic Information System (GIS) to realize the large-scale and inter-campuses equipment optimized assignment and information management. It discussed the mathematics model and the system architecture. Moreover, the paper described the key implementation technology in great detail such as spatial data mapping with MapInfo professional 9 and the development of WebGIS functions with MapXtreme. The results show that the solution is feasible and effective.

scholarly journals Efficient HPC Data Motion via Scratchpad Memory

2012 ◽  
Author(s):  
Kayla O Seager ◽  
Ananta Tiwari ◽  
Michael A. Laurenzano ◽  
Joshua Peraza ◽  
Pietro Cicotti ◽  
...  
Keyword(s):  
Scratchpad Memory

Streamlining Chinese Highway Accident Data Acquisition, Communications, and Analysis

2003 ◽  
Vol 1846 (1) ◽  
pp. 31-38 ◽  
Author(s):  
Ping Yi ◽  
Bin Ran
Keyword(s):  
Data Acquisition ◽  
Large Data ◽  
Highway Safety ◽  
Data Mapping ◽  
Data Logger ◽  
Safety Program ◽  
Crash Data ◽  
Accident Data ◽  
Analysis System ◽  
Set Up

This research examines a streamlined accident data acquisition, communications, and analysis system to improve the Chinese highway safety program. A data logger compatible with the Global Positioning System and geographic information system is proposed to identify highway accident locations and organize the data into a database format. A data encoding concept is used to transform Chinese characters into numbers, so that the encoded data are easy to integrate into a large data system. A three-tier client–server networking system is set up as the backbone framework for data communications between the central database and distributed local offices. Using local database functions, traffic police at the client level can view crash data through data mapping and attribute listing and analyze the data through nested query and sorting operations. A data graphing and analysis module was tested for automatically constructing a collision diagram on selected data. The proposed approach to crash data acquisition and analysis was found to be feasible and effective and will help to enhance China’s highway safety program after full implementation.

Sign in / Sign up

Export Citation Format

Share Document