r*g
2 楼
zero miss rate?
u r kidding. Do not you consider the cold start?
u r kidding. Do not you consider the cold start?
t*t
3 楼
given access pattern?
【在 t*******l 的大作中提到】
: open question.
【在 t*******l 的大作中提到】
: open question.
u*u
4 楼
sure, as long as you have developed Prescience capability through the use of
Melange.
【在 t*******l 的大作中提到】
: open question.
Melange.
【在 t*******l 的大作中提到】
: open question.
t*l
5 楼
considering the program is running for a long time and the miss rate is a
statistics, the initial fetch can be ignored.
【在 r********g 的大作中提到】
: zero miss rate?
: u r kidding. Do not you consider the cold start?
statistics, the initial fetch can be ignored.
【在 r********g 的大作中提到】
: zero miss rate?
: u r kidding. Do not you consider the cold start?
t*l
6 楼
assuming it is for video processing, so at least there are a lot of spatial
localitis for data cache.
for the instructions, let's also assume we have all the profiling statistics
and tracing capabilities, and we also have the power to change whatever we
want including the hardware, the cache's replacement policy, the OS, and
even the complier or how you code your program...
is it possible to achieve 0 miss rate in cache?
【在 t****t 的大作中提到】
: given access pattern?
localitis for data cache.
for the instructions, let's also assume we have all the profiling statistics
and tracing capabilities, and we also have the power to change whatever we
want including the hardware, the cache's replacement policy, the OS, and
even the complier or how you code your program...
is it possible to achieve 0 miss rate in cache?
【在 t****t 的大作中提到】
: given access pattern?
t*t
7 楼
i wouldn't say impossible, but surely it depends on your cache size vs data
size. I assume your 0 miss rate means for every data used, it's read from
memory once and only once.
Given an access pattern, you can calculate (or simulate) the minimum cache
size you need. For some deterministic case, you can get 0 miss rate. But
other than that, it really depends.
spatial
statistics
we
【在 t*******l 的大作中提到】
: assuming it is for video processing, so at least there are a lot of spatial
: localitis for data cache.
: for the instructions, let's also assume we have all the profiling statistics
: and tracing capabilities, and we also have the power to change whatever we
: want including the hardware, the cache's replacement policy, the OS, and
: even the complier or how you code your program...
: is it possible to achieve 0 miss rate in cache?
size. I assume your 0 miss rate means for every data used, it's read from
memory once and only once.
Given an access pattern, you can calculate (or simulate) the minimum cache
size you need. For some deterministic case, you can get 0 miss rate. But
other than that, it really depends.
spatial
statistics
we
【在 t*******l 的大作中提到】
: assuming it is for video processing, so at least there are a lot of spatial
: localitis for data cache.
: for the instructions, let's also assume we have all the profiling statistics
: and tracing capabilities, and we also have the power to change whatever we
: want including the hardware, the cache's replacement policy, the OS, and
: even the complier or how you code your program...
: is it possible to achieve 0 miss rate in cache?
相关阅读
C++ IDE and compiler关于 sorted and shifted 数组这个fast heapsort中文叫啥？两道M软件大公司的最新面世算法题 (转载)问个linux socket programming的问题请教，networking高手请问， 有这样的perl module 吗C语言一个passing variable的问题How can I write double data into file by C#?想学.net的同学看过来Looking for code example to get stock data这里有人用过mokaFive么？和VMWare什么区别？ (转载)问个perl Mechanize问题 how to do it ?=======Problem – coding in c++大家觉得Riverbed这家公司怎么样？ (转载)【请教】有很多contractor的组要不要去？ (转载)Help: undefined symbol你们都是用什么工具分析源代码(source code)?问个简单的C++问题Visual Fortran 6 和Windows Vista, 不兼容,怎么办?