Modern computing systems employ significant heterogeneity and specialization to meet
performance targets at manageable power. However, memory latency bottlenecks remain …

Current superscalar microprocessors' performance depends on its frequency and the
number of useful instructions that can be processed per cycle (IPC). In this paper we …

Decoupled architectures have not traditionally been used in the context of general purpose
computing because of their inability to tolerate control-intensive code that exists across a …

Speculative Precomputation (SP) is a technique to improve the latency of single-threaded
applications by utilizing idle multi-threading hardware resources to perform aggressive long …

The speed gap between processor and main memory is the major performance bottleneck of
modern computer systems. As a result, today's microprocessors suffer from frequent cache …

The decoupled access/execute architecture described a machine that enables the access of
memory values to be decoupled from the consumption of those values. Although never …

Computing systems have become ubiquitous in the modern world but their design is far from
one-size-fits-all. From battery-powered devices to supercomputers, deployment …

This work deals with the design, implementation and evaluation of a multithreading system
that enables fine-grained context switches without hardware support. The current chapter …

During the last ten years, multicore processors have matured from academic research
projects to real products in industry. They are now used in across almost the entire spectrum …

For years, single-thread performance was the most dominant force driving processor
development. In recent years, however, the poor scaling of single-thread super-scalar …