Heterogeneous multicore processors have emerged as an energy- and area-efficient architectural solution to improving performance for domain-specific applications such as those with a plethora of data-level parallelism. These processors typically contain a large number of small, compute-centric cores for acceleration while keeping one or two high-performance ILP cores on the die to guarantee single-thread performance. Although a major portion of the transistors are occupied by the acceleration cores, these resources will sit idle when running unparallelized legacy codes or the sequential part of an application. To address this underutilization issue, in this article, we introduce Chameleon, a flexible heterogeneous multicore architecture to virtualize these resources for enhancing memory performance when running sequential programs. The Chameleon architecture can dynamically virtualize the idle acceleration cores into a last-level cache, a data prefetcher, or a hybrid between these two techniques. In addition, Chameleon can operate in an adaptive mode that dynamically configures the acceleration cores between the hybrid mode and the prefetch-only mode by monitoring the effectiveness of the Chameleon cache mode. In our evaluation with SPEC2006 benchmark suite, different levels of performance improvements were achieved in different modes for different applications. In the case of the adaptive mode, Chameleon improves the performance of SPECint06 and SPECfp06 by 31% and 15%, on average. When considering only memory-intensive applications, Chameleon improves the system performance by 50% and 26% for SPECint06 and SPECfp06, respectively.

Several recent parallel computers have implemented logically shared, physically distributed memory systems which allow processors to directly access memory in other processors without interrupting the referenced processing element. Because this kind of architecture provides greater flexibility for interprocessor communications than private address space computers, different software models can be developed to take advantage of these machines. In this paper, we describe a low-level collective communications interface called SLICC (Shared address-space Library Interface for Collective Communications), that is specifically tailored for shared memory parallel computers. We also provide performance results for an implementation on the Cray T3D and compare its performance to Cray's PVM implementation. These results show that SLICC can be more than 10 times faster than PVM on the T3D.>