Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads

Kazuichi Oe, Mitsuru Sato, Takeshi Nanri

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

2 Citations (Scopus)

Abstract

The response time of solid state drives (SSDs) has dramatically reduced according to the spread of non-volatile memory express (NVMe) devices. These devices have response times of less than 100 micro seconds on average. The response time of all-flash-array systems has also drastically reduced through the use of NVMe SSDs. However, there are applications, particularly, virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response time. Their workloads were found to contain many input-output (IO) concentrations. We define IO concentration by using a declarative style. Input-output (IO) concentrations are aggregations of IO accesses. They appear in narrow regions of the storage volume and continue for periods of up to about an hour. These narrow regions occupy a few percent of the logical unit number capacity, include most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response time of these workloads, we developed automated tiered storage system called 'automated tiered storage with fast memory and slow flash storage' (ATSMF). The memory component of ATSMF is a memory with a non-volatile feature. The system predicts the remaining duration of IO concentration, calculates the response-time increase during migration and response-time decrease after migration, and migrates the IO concentrations if the response-time decrease after migration surpasses the response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and its memory access ratio is more than 50%.

Original languageEnglish
Title of host publicationProceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages311-317
Number of pages7
ISBN (Electronic)9781538620878
DOIs
Publication statusPublished - Apr 23 2018
Event5th International Symposium on Computing and Networking, CANDAR 2017 - Aomori, Japan
Duration: Nov 19 2017Nov 22 2017

Publication series

NameProceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017
Volume2018-January

Other

Other5th International Symposium on Computing and Networking, CANDAR 2017
CountryJapan
CityAomori
Period11/19/1711/22/17

Fingerprint

Response time (computer systems)
Data storage equipment
Agglomeration

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications
  • Hardware and Architecture

Cite this

Oe, K., Sato, M., & Nanri, T. (2018). Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads. In Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017 (pp. 311-317). (Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CANDAR.2017.25

Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads. / Oe, Kazuichi; Sato, Mitsuru; Nanri, Takeshi.

Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017. Institute of Electrical and Electronics Engineers Inc., 2018. p. 311-317 (Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017; Vol. 2018-January).

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

Oe, K, Sato, M & Nanri, T 2018, Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads. in Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017. Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 311-317, 5th International Symposium on Computing and Networking, CANDAR 2017, Aomori, Japan, 11/19/17. https://doi.org/10.1109/CANDAR.2017.25
Oe K, Sato M, Nanri T. Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads. In Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017. Institute of Electrical and Electronics Engineers Inc. 2018. p. 311-317. (Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017). https://doi.org/10.1109/CANDAR.2017.25
Oe, Kazuichi ; Sato, Mitsuru ; Nanri, Takeshi. / Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads. Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 311-317 (Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017).
@inproceedings{6ea28d7f49c64032b2e13d6fb21c2cdb,
title = "Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads",
abstract = "The response time of solid state drives (SSDs) has dramatically reduced according to the spread of non-volatile memory express (NVMe) devices. These devices have response times of less than 100 micro seconds on average. The response time of all-flash-array systems has also drastically reduced through the use of NVMe SSDs. However, there are applications, particularly, virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response time. Their workloads were found to contain many input-output (IO) concentrations. We define IO concentration by using a declarative style. Input-output (IO) concentrations are aggregations of IO accesses. They appear in narrow regions of the storage volume and continue for periods of up to about an hour. These narrow regions occupy a few percent of the logical unit number capacity, include most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response time of these workloads, we developed automated tiered storage system called 'automated tiered storage with fast memory and slow flash storage' (ATSMF). The memory component of ATSMF is a memory with a non-volatile feature. The system predicts the remaining duration of IO concentration, calculates the response-time increase during migration and response-time decrease after migration, and migrates the IO concentrations if the response-time decrease after migration surpasses the response-time increase during migration. Experimental results indicate that ATSMF is at least 20{\%} faster than flash storage only and its memory access ratio is more than 50{\%}.",
author = "Kazuichi Oe and Mitsuru Sato and Takeshi Nanri",
year = "2018",
month = "4",
day = "23",
doi = "10.1109/CANDAR.2017.25",
language = "English",
series = "Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "311--317",
booktitle = "Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017",
address = "United States",

}

TY - GEN

T1 - Automated Tiered Storage System Consisting of Memory and Flash Storage to Improve Response Time with Input-Output (IO) Concentration Workloads

AU - Oe, Kazuichi

AU - Sato, Mitsuru

AU - Nanri, Takeshi

PY - 2018/4/23

Y1 - 2018/4/23

N2 - The response time of solid state drives (SSDs) has dramatically reduced according to the spread of non-volatile memory express (NVMe) devices. These devices have response times of less than 100 micro seconds on average. The response time of all-flash-array systems has also drastically reduced through the use of NVMe SSDs. However, there are applications, particularly, virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response time. Their workloads were found to contain many input-output (IO) concentrations. We define IO concentration by using a declarative style. Input-output (IO) concentrations are aggregations of IO accesses. They appear in narrow regions of the storage volume and continue for periods of up to about an hour. These narrow regions occupy a few percent of the logical unit number capacity, include most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response time of these workloads, we developed automated tiered storage system called 'automated tiered storage with fast memory and slow flash storage' (ATSMF). The memory component of ATSMF is a memory with a non-volatile feature. The system predicts the remaining duration of IO concentration, calculates the response-time increase during migration and response-time decrease after migration, and migrates the IO concentrations if the response-time decrease after migration surpasses the response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and its memory access ratio is more than 50%.

AB - The response time of solid state drives (SSDs) has dramatically reduced according to the spread of non-volatile memory express (NVMe) devices. These devices have response times of less than 100 micro seconds on average. The response time of all-flash-array systems has also drastically reduced through the use of NVMe SSDs. However, there are applications, particularly, virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response time. Their workloads were found to contain many input-output (IO) concentrations. We define IO concentration by using a declarative style. Input-output (IO) concentrations are aggregations of IO accesses. They appear in narrow regions of the storage volume and continue for periods of up to about an hour. These narrow regions occupy a few percent of the logical unit number capacity, include most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response time of these workloads, we developed automated tiered storage system called 'automated tiered storage with fast memory and slow flash storage' (ATSMF). The memory component of ATSMF is a memory with a non-volatile feature. The system predicts the remaining duration of IO concentration, calculates the response-time increase during migration and response-time decrease after migration, and migrates the IO concentrations if the response-time decrease after migration surpasses the response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and its memory access ratio is more than 50%.

UR - http://www.scopus.com/inward/record.url?scp=85050221865&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050221865&partnerID=8YFLogxK

U2 - 10.1109/CANDAR.2017.25

DO - 10.1109/CANDAR.2017.25

M3 - Conference contribution

AN - SCOPUS:85050221865

T3 - Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017

SP - 311

EP - 317

BT - Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -