A tremendous amount of compiler research effort over the past ten years has been devoted to compensating for the deficiencies in hardware cache memory hierarchy. In this thesis, we propose a refocus of these energies towards compilation for memory hierarchy which is exposed to the compiler. With software exposure of memory hierarchy, not only can replacement policies be tailored to the application, but pollution can be minimized or elim-inated, allowing programs to achieve memory reference locality with less local memory than would be needed by a cache. Additionally, prefetch latencies are fully exposed and can be hidden by the compiler, thereby improving overall performance. We have developed a technique we call Compiler Controlled Hierarc...
Ease of programming is one of the main impediments for the broad acceptance of multi-core systems wi...
While CPU speed has been improved by a factor of 6400 over the past twenty years, memory bandwidth h...
Abstract — In many computer systems, a large portion of the execution time and energy consumption is...
The trend in high-performance microprocessor design is toward increasing computational power on the ...
The trend in high-performance microprocessor design is toward increasing computational power on the ...
Many applications are memory intensive and thus are bounded by memory latency and bandwidth. While i...
The performance of the memory hierarchy has become one of the most critical elements in the performa...
The gap between processors and main memory performance increases every year. In order to overcome th...
Modern microprocessor designs continue to obtain impressive performance gains through increasing clo...
Modern microprocessor designs continue to obtain impressive per-formance gains through increasing cl...
In order to mitigate the impact of the constantly widening gap between processor speed and main memo...
An ideal high performance computer includes a fast processor and a multi-million byte memory of comp...
This work was also published as a Rice University thesis/dissertation: http://hdl.handle.net/1911/19...
One of the primary challenges in embedded system design is designing the memory hierarchy and restru...
The potential of high-performance systems, especially parallel machines, is generally limited by the...
Ease of programming is one of the main impediments for the broad acceptance of multi-core systems wi...
While CPU speed has been improved by a factor of 6400 over the past twenty years, memory bandwidth h...
Abstract — In many computer systems, a large portion of the execution time and energy consumption is...
The trend in high-performance microprocessor design is toward increasing computational power on the ...
The trend in high-performance microprocessor design is toward increasing computational power on the ...
Many applications are memory intensive and thus are bounded by memory latency and bandwidth. While i...
The performance of the memory hierarchy has become one of the most critical elements in the performa...
The gap between processors and main memory performance increases every year. In order to overcome th...
Modern microprocessor designs continue to obtain impressive performance gains through increasing clo...
Modern microprocessor designs continue to obtain impressive per-formance gains through increasing cl...
In order to mitigate the impact of the constantly widening gap between processor speed and main memo...
An ideal high performance computer includes a fast processor and a multi-million byte memory of comp...
This work was also published as a Rice University thesis/dissertation: http://hdl.handle.net/1911/19...
One of the primary challenges in embedded system design is designing the memory hierarchy and restru...
The potential of high-performance systems, especially parallel machines, is generally limited by the...
Ease of programming is one of the main impediments for the broad acceptance of multi-core systems wi...
While CPU speed has been improved by a factor of 6400 over the past twenty years, memory bandwidth h...
Abstract — In many computer systems, a large portion of the execution time and energy consumption is...