Page-mapping techniques to reduce cache conflicts on CC-NUMA multiprocessors
نویسندگان
چکیده
Page-coloring and bin-hopping are two well-known page mapping schemes for reducing cache connicts. Previous work found bin-hopping to have 4% less cache miss rate than page-coloring on uniprocessor machines. Using execution-driven simulations, we nd that bin-hopping signiicantly outperforms simplistic page-coloring on CC-NUMA multipro-cessors. In certain cases, bin-hopping has 32% to 58% less execution time and over 60% fewer cache misses. By using part of the memory ID bits to hash the page color during page-mapping, we improve the performance of page-coloring to match that of bin-hopping on CC-NUMA multiprocessors.
منابع مشابه
ASCOMA: An Adaptive Hybrid Shared Memory Architecture
Scalable shared memory multiprocessors traditionally use either a cache coherent non uniform memory access CC NUMA or simple cache only memory architecture S COMA memory architecture Recently hybrid architectures that combine aspects of both CC NUMA and S COMA have emerged In this paper we present two improvements over other hybrid architectures The rst improvement is a page allocation algorith...
متن کاملDesign and Evaluation of a Switch Cache Architecture for CC-NUMA Multiprocessors
ÐCache coherent nonuniform memory access (CC-NUMA) multiprocessors provide a scalable design for shared memory. But, they continue to suffer from large remote memory access latencies due to comparatively slow memory technology and large data transfer latencies in the interconnection network. In this paper, we propose a novel hardware caching technique, called switch cache, to improve the remote...
متن کاملPerformance Evaluation of Memory Allocation Schemes on CC-NUMA Multiprocessors
{ Cache Coherent Non-Uniform Memory Access (CC-NUMA) architectures have received strong interests from both academia and industries. This paper studies the performance impact of design choices at diierent levels of address and memory mapping on CC-NUMA architectures. Through execution-driven simulations of ve numerical programs, we nd close interactions between data allocation, global address t...
متن کاملDesign and Evaluation of a Switch
Cache coherent non-uniform memory access (CC-NUMA) multiprocessors provide a scal-able design for shared memory but they continue to suuer from large remote memory access latencies due to comparatively slow memory technology and data transfer latencies in the in-terconnection network. In this paper, we propose a novel hardware caching technique, called switch cache, to improve the remote memory...
متن کاملSwitch Cache: A Framework for Improving the Remote Memory Access Latency of CC-NUMA Multiprocessors
Cache coherentnon-uniform memory access (CC-NUMA) multiprocessors continue to suffer from remote memory access latencies due to comparatively slow memory technology and data transfer latencies in the interconnection network. In this paper, we propose a novel hardware caching technique, called switch cache. The main idea is to implement small fast caches in crossbar switches of the interconnect ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Microprocessors and Microsystems - Embedded Hardware Design
دوره 22 شماره
صفحات -
تاریخ انتشار 1998