Gpm: leveraging persistent memory from a gpu authors : shweta pandey , aditya k kamath , arkaprava basu authors info & claims asplos '22: proceedings of the 27th acm international conference on architectural support for programming languages and operating systems. Gpm: leveraging persistent memory from a gpu authors: shweta pandey*, aditya k kamath*, arkaprava basu conference: acm conference on architectural support fo.
Gpm utilizes gpu’s parallelism to write millions of new rows and persist them in parallel from the kernel. in contrast, cap de signs are unable to leverage gpu’s parallelism as it persists from the cpu. further, initializing the dma engine and transferring rows from gpu to cpu memory adds overheads. This is a workshop presentation of ‘gpm’ published in asplos ‘23. this paper explored how to utilise commercially available nvm on a gpu using real hardware. through this process we came up with a benchmark suite (gpmbench) consisting of gpu applications that benefit from both gpu parallelism as well as nvm persistence. we also provide a gpu optimised library (libgpm) that simplifies. Gpm: leveraging persistent memory from a gpu shweta pandey*, aditya k kamath*, and arkaprava basu (*authors contributed equally to this work) published in acm 27th international conference on architectural support for programming languages and operating systems (asplos), 2022. Gpm: g pu with p ersistent m emory. gpm is a system which allows a gpu to leverage persistent memory and enables writing highly performant recoverable gpu applications. the repository contains the source of our benchmark suite: gpmbench and a cuda library: libgpm. gpmbench comprises of 9 benchmarks categorized as transactional, native and.
Gpm: leveraging persistent memory from a gpu shweta pandey*, aditya k kamath*, and arkaprava basu (*authors contributed equally to this work) published in acm 27th international conference on architectural support for programming languages and operating systems (asplos), 2022. Gpm: g pu with p ersistent m emory. gpm is a system which allows a gpu to leverage persistent memory and enables writing highly performant recoverable gpu applications. the repository contains the source of our benchmark suite: gpmbench and a cuda library: libgpm. gpmbench comprises of 9 benchmarks categorized as transactional, native and. Gpm’s potential to directly write and persist data. citation of original paper: shweta pandey, aditya k ka math, and arkaprava basu. gpm: leveraging persistent mem ory from a gpu. in proceedings of the 27th acm international conference on architectural support for programming lan guages and operating systems (asplos 2022). We take three key steps toward enabling gpu programs to access pm directly. first, enable direct access to pm from within a gpu kernel without needing to modify the hardware. next, we demonstrate three classes of gpu accelerated applications that benefit from pm. in the process, we create a workload suite with nine such applications.
Gpm’s potential to directly write and persist data. citation of original paper: shweta pandey, aditya k ka math, and arkaprava basu. gpm: leveraging persistent mem ory from a gpu. in proceedings of the 27th acm international conference on architectural support for programming lan guages and operating systems (asplos 2022). We take three key steps toward enabling gpu programs to access pm directly. first, enable direct access to pm from within a gpu kernel without needing to modify the hardware. next, we demonstrate three classes of gpu accelerated applications that benefit from pm. in the process, we create a workload suite with nine such applications.
Comments are closed.