Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection

Yu Omori, Keiji Kimura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The emerging technology of byte-Addressable nonvolatile memory chips is expected to enable larger main memory and lower power consumption than the traditional DRAM. It also realizes durable data structure without ordinary file systems. However, while enumerating the advantages of nonvolatile main memory (NVMM), its write-Time expensive latency and higher energy consumption in comparision with a DRAM must be considered. These special characteristics of NVMM require new compiler techniques and OS support as well as new memory architectures. Several NVMM emulators built on real machines have been proposed to facilitate those software and hardware researches. Their designs were originally based on a simple coarse-grain delay model that injected additional clock cycles in the read and write requests sent to the memory controller. However, they could not utilize bank-level parallelism and row-buffer access locality, relied on by today's memory modules, to exploit their performance. Therefore, a fine-grain delay model was recently proposed where the delay is injected for the primitive memory operations issued by the memory controller. In this paper, we implement both the coarse-grain and the fine-grain delay models on an SoC-FPGA board along with the use of Linux kernel modifications and several runtime functions. Then, the program behavior differences between two models are evaluated with SPEC CPU programs. The fine-grain model reveals the program execution time is influenced by the frequency of NVMM memory requests rather than the cache hit ratio. Bank-level parallelism and row-buffer access locality also affect the memory access delay, and the fine-grain model shows lower execution time for four of fourteen programs than the coarse-grain even when the former has longer total write latency.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728138541
DOIs
Publication statusPublished - 2019 Aug
Event8th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019 - Hangzhou, China
Duration: 2019 Aug 182019 Aug 21

Publication series

NameProceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019

Conference

Conference8th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019
CountryChina
CityHangzhou
Period19/8/1819/8/21

Fingerprint

Data storage equipment
Dynamic random access storage
Memory architecture
Controllers
Program processors
Data structures
Field programmable gate arrays (FPGA)
Clocks
Electric power utilization
Energy utilization
Hardware

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Hardware and Architecture

Cite this

Omori, Y., & Kimura, K. (2019). Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection. In Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019 [8863522] (Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NVMSA.2019.8863522

Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection. / Omori, Yu; Kimura, Keiji.

Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8863522 (Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Omori, Y & Kimura, K 2019, Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection. in Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019., 8863522, Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019, Institute of Electrical and Electronics Engineers Inc., 8th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019, Hangzhou, China, 19/8/18. https://doi.org/10.1109/NVMSA.2019.8863522
Omori Y, Kimura K. Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection. In Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 8863522. (Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019). https://doi.org/10.1109/NVMSA.2019.8863522
Omori, Yu ; Kimura, Keiji. / Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection. Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019. Institute of Electrical and Electronics Engineers Inc., 2019. (Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019).
@inproceedings{4ee5ebf2f3124917aa88cac8e2167884,
title = "Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection",
abstract = "The emerging technology of byte-Addressable nonvolatile memory chips is expected to enable larger main memory and lower power consumption than the traditional DRAM. It also realizes durable data structure without ordinary file systems. However, while enumerating the advantages of nonvolatile main memory (NVMM), its write-Time expensive latency and higher energy consumption in comparision with a DRAM must be considered. These special characteristics of NVMM require new compiler techniques and OS support as well as new memory architectures. Several NVMM emulators built on real machines have been proposed to facilitate those software and hardware researches. Their designs were originally based on a simple coarse-grain delay model that injected additional clock cycles in the read and write requests sent to the memory controller. However, they could not utilize bank-level parallelism and row-buffer access locality, relied on by today's memory modules, to exploit their performance. Therefore, a fine-grain delay model was recently proposed where the delay is injected for the primitive memory operations issued by the memory controller. In this paper, we implement both the coarse-grain and the fine-grain delay models on an SoC-FPGA board along with the use of Linux kernel modifications and several runtime functions. Then, the program behavior differences between two models are evaluated with SPEC CPU programs. The fine-grain model reveals the program execution time is influenced by the frequency of NVMM memory requests rather than the cache hit ratio. Bank-level parallelism and row-buffer access locality also affect the memory access delay, and the fine-grain model shows lower execution time for four of fourteen programs than the coarse-grain even when the former has longer total write latency.",
author = "Yu Omori and Keiji Kimura",
year = "2019",
month = "8",
doi = "10.1109/NVMSA.2019.8863522",
language = "English",
series = "Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019",

}

TY - GEN

T1 - Performance Evaluation on NVMM Emulator Employing Fine-Grain Delay Injection

AU - Omori, Yu

AU - Kimura, Keiji

PY - 2019/8

Y1 - 2019/8

N2 - The emerging technology of byte-Addressable nonvolatile memory chips is expected to enable larger main memory and lower power consumption than the traditional DRAM. It also realizes durable data structure without ordinary file systems. However, while enumerating the advantages of nonvolatile main memory (NVMM), its write-Time expensive latency and higher energy consumption in comparision with a DRAM must be considered. These special characteristics of NVMM require new compiler techniques and OS support as well as new memory architectures. Several NVMM emulators built on real machines have been proposed to facilitate those software and hardware researches. Their designs were originally based on a simple coarse-grain delay model that injected additional clock cycles in the read and write requests sent to the memory controller. However, they could not utilize bank-level parallelism and row-buffer access locality, relied on by today's memory modules, to exploit their performance. Therefore, a fine-grain delay model was recently proposed where the delay is injected for the primitive memory operations issued by the memory controller. In this paper, we implement both the coarse-grain and the fine-grain delay models on an SoC-FPGA board along with the use of Linux kernel modifications and several runtime functions. Then, the program behavior differences between two models are evaluated with SPEC CPU programs. The fine-grain model reveals the program execution time is influenced by the frequency of NVMM memory requests rather than the cache hit ratio. Bank-level parallelism and row-buffer access locality also affect the memory access delay, and the fine-grain model shows lower execution time for four of fourteen programs than the coarse-grain even when the former has longer total write latency.

AB - The emerging technology of byte-Addressable nonvolatile memory chips is expected to enable larger main memory and lower power consumption than the traditional DRAM. It also realizes durable data structure without ordinary file systems. However, while enumerating the advantages of nonvolatile main memory (NVMM), its write-Time expensive latency and higher energy consumption in comparision with a DRAM must be considered. These special characteristics of NVMM require new compiler techniques and OS support as well as new memory architectures. Several NVMM emulators built on real machines have been proposed to facilitate those software and hardware researches. Their designs were originally based on a simple coarse-grain delay model that injected additional clock cycles in the read and write requests sent to the memory controller. However, they could not utilize bank-level parallelism and row-buffer access locality, relied on by today's memory modules, to exploit their performance. Therefore, a fine-grain delay model was recently proposed where the delay is injected for the primitive memory operations issued by the memory controller. In this paper, we implement both the coarse-grain and the fine-grain delay models on an SoC-FPGA board along with the use of Linux kernel modifications and several runtime functions. Then, the program behavior differences between two models are evaluated with SPEC CPU programs. The fine-grain model reveals the program execution time is influenced by the frequency of NVMM memory requests rather than the cache hit ratio. Bank-level parallelism and row-buffer access locality also affect the memory access delay, and the fine-grain model shows lower execution time for four of fourteen programs than the coarse-grain even when the former has longer total write latency.

UR - http://www.scopus.com/inward/record.url?scp=85074147233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074147233&partnerID=8YFLogxK

U2 - 10.1109/NVMSA.2019.8863522

DO - 10.1109/NVMSA.2019.8863522

M3 - Conference contribution

AN - SCOPUS:85074147233

T3 - Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019

BT - Proceedings - 2019 IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -