Search

In multicore processors, a cache coherence protocol is used for maintaining the data coherence in private caches. Traditional snoop-based protocols used broadcasting messages for maintaining data coherence to result in many tag comparisons in private caches. However, the use of broadcasting messages consumes a considerable amount of energy as well as execution time because of tag comparisons. In this paper, an energy-efficient hybrid coherence protocol (EEHCP) is proposed to reduce energy consumption for maintaining the data coherence of private caches by reducing the amount of broadcasting messages and tag comparisons. According to the simulation results, the proposed EEHCP consumes 27, 18, and 46% less energy than the SFT (snoop filter table) protocol, the snoop-based protocol with the hybrid write strategy, and the traditional snoop-based protocol with MESI, and 18, 17, and 32% less execution time than these three protocols, respectively. [ABSTRACT FROM AUTHOR]

The cache memory consumes a large proportion of the energy used by a processor. In the on-chip cache, the translation lookaside buffer (TLB) accounts for 20–50% of energy consumption of the on-chip cache. To reduce energy consumption caused by TLB accesses, a virtual cache can be accessed by virtual addresses which are issued by a processor directly. However, a virtual cache may result in the synonym problem. In this paper, we propose low-cost synonym detection hardware and a synonym data coherence mechanism. These reduce the energy consumption incurred by TLB lookups, and maintain synonym data consistency in the virtual cache. The proposed synonym detection hardware efficiently reduces the number of blocks that must be looked up in a virtual cache for saving energy. In addition, the proposed synonym data coherence mechanism also reduces the number of invalidated blocks in the virtual cache to prevent the destruction of cache locality. The simulation results show that our proposed energy-aware virtual cache consumes 51%, 27%, and 20% less energy than the traditional physical cache, traditional virtual cache, and synonym lookaside buffer (SLB), respectively. In addition, our proposed design shows almost the same static energy consumption as SLB, and reduces static energy consumption by about 20% compared with the traditional physical cache and virtual cache. [ABSTRACT FROM AUTHOR]

Abstract: The power consumed by memory systems accounts for 45% of the total power consumed by an embedded system, and the power consumed during a memory access is 10 times higher than during a cache access. Thus, increasing the cache hit rate can effectively reduce the power consumption of the memory system and improve system performance. In this study, we increased the cache hit rate and reduced the cache-access power consumption by developing a new cache architecture known as a single linked cache (SLC) that stores frequently executed instructions. SLC has the features of low power consumption and low access delay, similar to a direct mapping cache, and a high cache hit rate similar to a two way-set associative cache by adding a new link field. In addition, we developed another design known as a multiple linked caches (MLC) to further reduce the power consumption during each cache access and avoid unnecessary cache accesses when the requested data is absent from the cache. In MLC, the linked cache is split into several small linked caches that store frequently executed instructions to reduce the power consumption during each access. To avoid unnecessary cache accesses when a requested instruction is not in the linked caches, the addresses of the frequently executed blocks are recorded in the branch target buffer (BTB). By consulting the BTB, a processor can access the memory to obtain the requested instruction directly if the instruction is not in the cache. In the simulation results, our method performed better than selective compression, traditional cache, and filter cache in terms of the cache hit rate, power consumption, and execution time. [Copyright &y& Elsevier]

The many-core SoC is a future trend technology, and the process yield will face many unpredictable challenges. Nonuniform cache architecture (NUCA) can improve the performance of many-core SoC for embedded systems. It embeds a NoC into the cache memory to enhance the data access by distributing traffic loads to several banks in parallel. Providing fault-tolerant mechanism in NUCA is very important because the chip can still work efficiently when some memory banks are unusable. In this paper, we design a specific router working with static and dynamic cache remapping policies to support faulty banks in NUCA. When a L2 cache bank in NUCA is unusable, static remapping policy (SRP) selects a suitable neighbor cache bank according to the collected remapping cost to assist with the cache access by considering cache status and traffic status of the router. We also propose a dynamic remapping policy (DRP) to select the suitable cache bank dynamically at runtime to fit the real loading status of neighbor nodes around the faulty bank. The experimental results show that the average improvement of the SRP is approximated to 26 %, and the average improvement of the DRP is approximated to 28 %. [ABSTRACT FROM AUTHOR]

In embedded systems, cache is commonly used to improve system performance. However, the cache consumes a large amount of power, and among the components of the cache memory, tag comparisons consume the most amount of power. Therefore, how to design a cache that does not consume so much power when comparing tags and that has a high hit ratio is an important challenge. In this paper, we propose a Tagless Instruction Cache, called TL-IC, that does not perform tag comparisons in order to save power in embedded systems. To guarantee that an instruction fetched from TL-IC is the desired instruction, instead of cache lines being used, the basic blocks of programs are placed into TL-IC. In addition, to utilize TL-IC as much as possible in order to save the most amount of power and to take into account the general-purpose and special-purpose applications, both the static allocation and the dynamic allocation of basic blocks are used to select the frequently executed basic blocks of programs in TL-IC. With a high utilization of TL-IC that does not perform tag comparisons, the power consumed in fetching instructions can be efficiently reduced. In the simulation results, we show and compare the power consumption of our proposed TL-IC, L0 cache, Linebuffer, and TH-IC. [ABSTRACT FROM AUTHOR]