Abstract-In this paper, we explore the possibility of using STT-RAM technology to completely replace DRAM in main memory. Our goal is to make STT-RAM performance comparable to DRAM while providing substantial power savings. Towards this goal, we first analyze the performance and energy of STT-RAM, and then identify key optimizations that can be employed to improve its characteristics. Specifically, using partial write and row buffer write bypass, we show that STT-RAM main memory performance and energy can be significantly improved. Our experiments indicate that an optimized, equal capacity STT-RAM main memory can provide performance comparable to DRAM main memory, with an average 60% reduction in main memory energy.

No abstract

Given the diverse range of application characteristics that chip multiprocessors (CMPs) need to cater to, a "onecache-topology-fits-all" design philosophy will clearly be inadequate. In this paper, we propose MorphCache, a Reconfigurable Adaptive Multi-level Cache hierarchy. MorphCache dynamically tunes a multi-level cache topology in a CMP to allow significantly different cache topologies to exist on the same architecture. Starting from per-core L2 and L3 cache slices as the basic design point, MorphCache alters the cache topology dynamically by merging or splitting cache slices and modifying the accessibility of different cache slice groups to different cores in a CMP. We evaluated MorphCache on a 16 core CMP on a full system simulator and found that it significantly improves both average throughput and harmonic mean of speedups of diverse multithreaded and multiprogrammed workloads. Specifically, our results show that MorphCache improves throughput of the multiprogrammed mixes by 29.9% over a topology with all-shared L2 and L3 caches and 27.9% over a topology with per core private L2 cache and shared L3 cache. In addition, we also compared MorphCache to partitioning a single shared cache at each level using promotion/insertion pseudo-partitioning (PIPP) [28] and managing per-core private cache at each level using dynamic spill receive caches (DSR) [18]. We found that MorphCache improves average throughput by 6.6% over PIPP and by 5.7% over DSR when applied to both L2 and L3 caches.

No abstract

Device level heterogeneity promises high energy efficiency over a larger range of voltages than a single device technology alone can provide. In this paper, starting from device models, we first present ground-up modeling of CMOS and TFET cores, and verify this model against existing processors. Using our core models, we construct a 32-core TFET-CMOS heterogeneous multicore. We then show that it is a very challenging task to identify the ideal runtime configuration to use in such a heterogeneous multicore, which includes finding the best number/type of cores to activate and the corresponding voltages/frequencies to select for these cores. In order to effectively utilize this heterogeneous processor, we propose a novel automated runtime scheme. Our scheme is designed to automatically improve the performance of applications running on heterogeneous CMOS-TFET multicores operating under a fixed power budget, without requiring any effort from the application programmer or the user. Our scheme combines heterogeneous thread-to-core mapping, dynamic work partitioning, and dynamic power partitioning to identify energy efficient operating points. With simulations we show that our runtime scheme can enable a CMOS-TFET multicore to serve a diversity of workloads with high energy efficiency and achieve 21% average speedup over the best performing equivalent homogeneous multicore.