Browse > Article
http://dx.doi.org/10.14372/IEMEK.2019.14.2.63

An Adaptive Polling Selection Technique for Ultra-Low Latency Storage Systems  

Chun, Myoungjun (Seoul National University)
Kim, Yoona (Seoul National University)
Kim, Jihong (Seoul National University)
Publication Information
IEMEK Journal of Embedded Systems and Applications / v.14, no.2, 2019 , pp. 63-69 More about this Journal
Abstract
Recently, ultra-low latency flash storage devices such as Z-SSD and Optane SSD were introduced with the significant technological improvement in the storage devices which provide much faster response time than today's other NVMe SSDs. With such ultra-low latency, $10{\mu}s$, storage devices the cost of context switch could be an overhead during interrupt-driven I/O completion process. As an interrupt-driven I/O completion process could bring an interrupt handling overhead, polling or hybrid-polling for the I/O completion is known to perform better. In this paper, we analyze tail latency problem in a polling process caused by process scheduling in data center environment where multiple applications run simultaneously under one system and we introduce our adaptive polling selection technique which dynamically selects efficient processing method between two techniques according to the system's conditions.
Keywords
Flash memory; Storage system; Polling; I/O stack optimization;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Samsung Z-SSD SZ985, Avaliable on : https://www.samsung.com/us/labs/pdfs/-collateral/Samsung_Z-NAND_Technology_Brief_v5.pdf.
2 Intel Optane SSD 900P, Avaliable on : https://www.intel.com/content/www/us/en/-products-memory-storage/solid-state-drives/gaming-enthusiast-ssds/optane-900p-series.html.
3 J. Yang, D.B. Minturn, F. Hady, "When Poll is Better than Interrupt," Proceedings of the USENIX Conference on File and Storage Technologies, Vol. 12, pp. 3-3, 2012.
4 D.L. Moal., "I/O Latency Optimization with Polling," Proceedings of Valut-Linux Storage and Filesystems Conference, 2012.
5 A. Eisenman, D. Gardner, I.. AbdellRahman, J. Axboe, S. Dong, K. Hazelwood, C. Petersen, A. Cidon, S. Katti, "Reducing DRAM Footprint with NVM in Facebook", Proceedings of the Europeam Conference on Computer Systems, 2018.
6 J. Zhang, M. Kwon, D. Gouk, S. Koh, C. Lee, M. Alian, M. Chun, M.T. Kandemir, N. Kim, M. Jung, "FlashShare: Punching Through Server Storage Stack from Kernel to Firmware for Ultra-Low Lantecy SSDs", Proceedings of the USENIX Symposium on Operating Systems Design and Implementation, pp. 477-492, 2018.
7 U. Kang, C.E. Tsourakakis, C. Faloutsos, "Pegasus: A Peta-Scale Graph Mining System", Proceedings of IEEE International Conference on Data Mining, pp. 229-238, 2009.
8 J. Dean, L. Barroso, "The Tail at Scale", Journal of Communications of the ACM, Vol. 56, No. 2, pp. 74-80, 2013.   DOI
9 Flexible I/O Tester, Avaliable on : https://github.com/axboe/fio.