Energy-Efficient Cache Coherence for Embedded Multi-Processor Systems through Application-Driven Snoop Filtering

Abstract

We present a novel methodology for power reduction in embedded multiprocessor systems. Maintaining local caches coherent in bus-based multiprocessor systems results in significantly elevated power consumption, as the bus snooping protocols result in local cache lookups for each memory reference placed on the common bus. Such a conservative approach is warranted in general-purpose systems, where no prior knowledge regarding the communication structure between threads or processes is available. In such a general-purpose context the assumption is that each memory request is potentially a reference to a shared memory region, which may result in cache inconsistency, if no correcting activities are undertaken. The approach we propose exploits the fact that in embedded systems, important knowledge is available to the system designers regarding communication activities between tasks allocated to the different processor nodes. We demonstrate how the snoop-related cache probing activity can be drastically reduced by identifying in a deterministic way all the shared memory regions and the communication patterns between the processor nodes. Cache snoop activity is enabled only for the fraction of the bus transactions, which refer to locations belonging to known shared memory region for each processor node; for the remaining larger part of memory references known to be of no relation to the given processor node, snoop probings in the local cache are completely disabled, thus saving a large amount of power. The required hardware support is not only cost-efficient, but is also software programmable, which allows the system software to dynamically customize the cache coherence controller to the needs of different tasks or even different parts of the same program. The experiments which we have performed on a number of important applications demonstrate the effectiveness of the proposed approach.