Virtually Splitting Data Caches using Multiple Address Decoders
نویسندگان
چکیده
An address decoder is a small hardware unit that uses an address to index and place the data into memory units including cache memories. In current CPU cache designs there is a single decoder unit which serves to place data into the cache. In this paper we describe a technique to reduce contention on CPU’s caches through the use of multiple address decoders. We argue that by using multiple decoding techniques better data placement can be achieved and the CPU cache can be better utilized. We present an overview of an instrumentation tool developed to collect fine-grained data traces and a technique for virtually splitting caches using separate address decoders. Our results demonstrate the feasibility and the impact of virtual cache splitting.
منابع مشابه
Leakage Power Optimization Techniques for Ultra Deep Sub - Micron Multi - Level Caches Nam
On-chip L1 and L2 caches represent a sizeable fraction of the total power consumption of microprocessors. In deep sub-micron technology , the subthreshold leakage power is becoming the dominant fraction of the total power consumption of those caches. In this paper, we present optimization techniques to reduce the leakage power of on-chip caches assuming that there are multiple threshold voltage...
متن کاملTime-Splitting Multiple-Access
It is shown that the encoding/decoding problem for any asynchronous M -user memoryless multiple-access channel can be reduced to corresponding problems for at most 2M 1 single-user memoryless channels. This is done via a method called time-splitting multiple-access which is closely related to a recently developed method called rate-splitting multiple access. It is also related to to multilevel ...
متن کاملA Quantitative Evaluation of Cache Types for High-Performance Computer Systems
Parallel accesses to the table lookaside buffer (TLB) and cache array are crucial for high-performance computer systems, and the choice of cache types is one of the most important factors affecting cache performance. In this paper we classify caches according to both index and tag. Since both index and tag could be either virtual (V) or real (R), our classification results in four combinations ...
متن کاملReducing Address Translation Overheads with Virtual Caching
This dissertation research addresses overheads in supporting virtual memory, especially virtual-to-physical address translation overheads (i.e., performance, power, and energy) via a Translation Lookaside Buffer (TLB). To overcome the overheads, we revisit virtually indexed, virtually tagged caches. In practice, they have not been common in commercial microarchitecture designs, and the crux of ...
متن کاملA memory management unit and cache controller for the MARS systeiil
For large caches, the interaction between cache access and address translation affects the machine cycle time and the access time to memory. The physically addressed caches slow down the cache access due to the virtual address translation. The virtually addressed caches is faster, but the synonym problem is difficult to handle. By some software constraints and hardware support, our virtually ad...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2012