Recent years have witnessed phenomenal growth in the computational capabilities and
applications of GPUs. However, this trend has also led to a dramatic increase in their power …

Process variation—deviation in parameters from their nominal specifications—threatens to
slow down and even pause technological scaling, and mitigation of it is the way to continue …

This paper presents Warped-Compression, a warp-level register compression scheme for
reducing GPU power consumption. This work is motivated by the observation that the …

SRAM based register file (RF) is one of the major factors limiting the scaling of GPGPU. In
this work, we propose to use the emerging nonvolatile domain-wall-shift-write based …

Graphics Processing Units (GPUs) employ large register files to accommodate all active
threads and accelerate context switching. Unfortunately, register files are a scalability …

We study adding aggregate operators, such as summing up elements of a column of a
relation, to logics with counting mechanisms. The primary motivation comes from database …

This paper presents a pre-execution approach for improving GPU performance, called P-
mode (pre-execution mode). GPUs utilize a number of concurrent threads for hiding …

To support their massively-multithreaded architecture, GPUs use very large register file (RF)
which has a capacity higher than even L1 and L2 caches. In total contrast, traditional CPUs …

Cryogenic computing, which is to run a computer at extremely low temperatures (eg, 77K), is
a highly promising solution to dramatically improve the computer's performance and power …

The register file is one of the largest and most power-hungry structures in a Graphics
Processing Unit (GPU), because massive multithreading requires all the register state for …