Unblinding the OS to optimize user-perceived flash SSD latency

Woong Shin; Jaehyun Park; Heon Y. Yeom

Unblinding the OS to optimize user-perceived flash SSD latency

Woong Shin, Jaehyun Park, Heon Y. Yeom

Research output: Contribution to conference › Paper › peer-review

Abstract

In this paper, we present a flash solid-state drive (SSD) optimization that provides hints of SSD internal behaviors, such as device I/O time and buffer activities, to the OS in order to mitigate the impact of I/O completion scheduling delays. The hints enable the OS to make reliable latency predictions of each I/O request so that the OS can make accurate scheduling decisions when to yield or block (busy wait) the CPU, ultimately improving user-perceived I/O performance. This was achieved by implementing latency predictors supported with an SSD I/O behavior tracker within the SSD that tracks I/O behavior at the level of internal resources, such as DRAM buffers or NAND chips. Evaluations with an SSD prototype based on a Xilinx Zynq-7000 FPGA and MLC flash chips showed that our optimizations enabled the OS to mask the scheduling delays without severely impacting system parallelism compared to prior I/O completion methods.

Original language	English
State	Published - 2016
Externally published	Yes
Event	8th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2016, co-located with USENIX ATC 2016 - Denver, United States Duration: Jun 20 2016 → Jun 21 2016

Conference

Conference	8th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2016, co-located with USENIX ATC 2016
Country/Territory	United States
City	Denver
Period	06/20/16 → 06/21/16

Funding

We would like to thank the anonymous USENIX Hot-Storage reviewers. This research was supported by Next-Generation Information Computing Development Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Science, ICT & Future Plannig (2015M 3C 4A7065646).

Funders	Funder number
Ministry of Science, ICT & Future Plannig	2015M 3C 4A7065646
National Research Foundation of Korea

Cite this

@conference{a22e0bfad9304c00bdb00806c1e6f373,

title = "Unblinding the OS to optimize user-perceived flash SSD latency",

abstract = "In this paper, we present a flash solid-state drive (SSD) optimization that provides hints of SSD internal behaviors, such as device I/O time and buffer activities, to the OS in order to mitigate the impact of I/O completion scheduling delays. The hints enable the OS to make reliable latency predictions of each I/O request so that the OS can make accurate scheduling decisions when to yield or block (busy wait) the CPU, ultimately improving user-perceived I/O performance. This was achieved by implementing latency predictors supported with an SSD I/O behavior tracker within the SSD that tracks I/O behavior at the level of internal resources, such as DRAM buffers or NAND chips. Evaluations with an SSD prototype based on a Xilinx Zynq-7000 FPGA and MLC flash chips showed that our optimizations enabled the OS to mask the scheduling delays without severely impacting system parallelism compared to prior I/O completion methods.",

author = "Woong Shin and Jaehyun Park and Yeom, {Heon Y.}",

note = "Publisher Copyright: {\textcopyright} USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2016, co-located with USENIX ATC 2016.All right reserved.; 8th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2016, co-located with USENIX ATC 2016 ; Conference date: 20-06-2016 Through 21-06-2016",

year = "2016",

language = "English",

}

TY - CONF

T1 - Unblinding the OS to optimize user-perceived flash SSD latency

AU - Shin, Woong

AU - Park, Jaehyun

AU - Yeom, Heon Y.

PY - 2016

Y1 - 2016

N2 - In this paper, we present a flash solid-state drive (SSD) optimization that provides hints of SSD internal behaviors, such as device I/O time and buffer activities, to the OS in order to mitigate the impact of I/O completion scheduling delays. The hints enable the OS to make reliable latency predictions of each I/O request so that the OS can make accurate scheduling decisions when to yield or block (busy wait) the CPU, ultimately improving user-perceived I/O performance. This was achieved by implementing latency predictors supported with an SSD I/O behavior tracker within the SSD that tracks I/O behavior at the level of internal resources, such as DRAM buffers or NAND chips. Evaluations with an SSD prototype based on a Xilinx Zynq-7000 FPGA and MLC flash chips showed that our optimizations enabled the OS to mask the scheduling delays without severely impacting system parallelism compared to prior I/O completion methods.

AB - In this paper, we present a flash solid-state drive (SSD) optimization that provides hints of SSD internal behaviors, such as device I/O time and buffer activities, to the OS in order to mitigate the impact of I/O completion scheduling delays. The hints enable the OS to make reliable latency predictions of each I/O request so that the OS can make accurate scheduling decisions when to yield or block (busy wait) the CPU, ultimately improving user-perceived I/O performance. This was achieved by implementing latency predictors supported with an SSD I/O behavior tracker within the SSD that tracks I/O behavior at the level of internal resources, such as DRAM buffers or NAND chips. Evaluations with an SSD prototype based on a Xilinx Zynq-7000 FPGA and MLC flash chips showed that our optimizations enabled the OS to mask the scheduling delays without severely impacting system parallelism compared to prior I/O completion methods.

UR - http://www.scopus.com/inward/record.url?scp=85088230508&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85088230508

T2 - 8th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2016, co-located with USENIX ATC 2016

Y2 - 20 June 2016 through 21 June 2016

ER -

Unblinding the OS to optimize user-perceived flash SSD latency

Abstract

Conference

Funding

Other files and links

Fingerprint

Cite this