DEIM Forum 2014 D3-5 DSMS DSMS DSMS 2.13% RTOS Realtime-Aware Efficient Query Processing for Automotiv

Size: px

Start display at page:

Download "DEIM Forum 2014 D3-5 DSMS DSMS DSMS 2.13% RTOS Realtime-Aware Efficient Query Processing for Automotiv"

あきみとくやす
5 years ago
Views:

1 DEIM Forum 2014 D3-5 DSMS DSMS DSMS 2.13% RTOS Realtime-Aware Efficient Query Processing for Automotive DSMS Satoshi KATSUNUMA, Shinya HONDA, and Hiroaki TAKADA Graduate School of Information Science, Nagoya University Furo-cho, Chikusa-ku, Nagoya, Japan Central Research Lab., Hitachi Ltd AUTOSAR AUTOSAR ECU PF [1] [2] [3] AU- TOSAR ECU PF DSMS ECU DSMS [7] [8] DSMS RTOS RTOS

2 2 DSMS DSMS DSMS 2. 1 RTOS (a) ECU RTOS 12 RTOS ECU RTOS RTOS RTOS (100ms) () : (100ms) () (100ms) () 1 : 0ms 50ms 100ms 150ms 200ms 250ms (a) (b) (c) 0ms 50ms 100ms 150ms 200ms 250ms / 0ms 50ms 100ms 150ms 200ms 250ms 2. 2 DSMS PF DSMS DSMSeDSMS[2] [4] DSMS [7] [8] [8]

3 2 DSMS (b) 50ms 1(b1) RTOS 1 1(c) 2 1(c1) RTOS 3a edsms 3b 3c edsms [4] Borealis [10] edsms edsms edsms RTOS 3a1 DSMS 3b1 DSMS 3b2 3b3 3b5

4 (b1)dsms (b2) (b2) (b1) DSMS (b2) (b1) DSMS 3 α β γ DSMS DSMS DSMS α β γ a 3b6 3b7 3b8 3b4 DSMS 3b (c1) c 100ms 50ms 4 DSMS RTOS 41 RTOS DSMS DSMS 41 DSMS 4(4) a b DSMS DSMS 4(2) 4(3) DSMS

5 b b b 50km/h -> 60km/h DSMS b b b 50km/h -> 60km/h 5 DSMS a b 60km/h a b 60km/h 4(2) a b DSMS 43 a a edsms [4] b b b b 52 b (c2) A B (A)(B) (1)(3) 6(1) a 6(2) 6(3) a b b 65 b DSMS 66

6 b b b 50km/h -> 60km/h DSMS a b 60km/h b b b 50km/h -> 60km/h DSMS a b 60km/h α β γ b b b 50km/h -> 60km/h b DSMS DSMS DSMS a b 60km/h 6 DSMS 4. 5 Non-preemptive section Non-preemptive section(nps) NPS 7 71 NPS a NPS 72 a a 73 b NPS RTOS NPS NPS NPS NPS NPS ZMP RoboCar [9] 8

7 DSMS DSMS a Non-preemptive section a 7 DSMS b α β γ 60km/h Non-preemptive section 8 50ms20ms ms50ms (b) 50ms NPS NPS 1 NPS 0 us 280 us 13,100 us 0 us 13,380 us 13,100 us 13,100 us 26,200 us NPS NPS 1(b1) 90 13,100us 13,100us NPS NPS NPS

8 280us NPS NPS 280us 280us 2.13% RTOS Query Processing, Resource Management, and Approximation in a Data Stream Management System, Conference on Innovative Data Systems Research (CIDR), [6] D. J. Abadi, D. Carney, U. Cetintemel, M. Cherniack, C. Convey, S. Lee, M. Stonebraker, N. Tatbul, and S. Zdonik: Aurora: a new model and architecture for data stream management, The VLDB Journal, [7] J. Chen, D. J. DeWitt, and J. F. Naughton: Design and Evaluation of Alternative Selection Placement Strategies in Optimizing Continuous Queries, ICDE [8], :, Vol. J87-D-I, No. 10, pp , [9] ZMP: RoboCarR 1/10 / ZMP RC-Z, [10] Borealis Distributed Stream Processing Engine, % [1] : Vol.42, No.2, [2] : Vol. J95-D, No.12, pp , Dec, 2012 [3],,, :, 11, Vol.2, pp.7-14, [4] : TODvol.55, [5] R. Motwani, J. WidomA. Arasu, B. Babcock, S. Babu, M. Datar, G. Manku, C. Olston, J. Rosenstein, and R. Varma:

DEIM Forum 2015 E4-5 DSMS DSMS DSMS 32% 46% RTOS Priority Inversion Time

DEIM Forum 2015 E4-5 DSMS DSMS DSMS 32% 46% RTOS Priority Inversion Time DEIM Forum 2015 E4-5 DSMS 464 8601 E-mail: {katsunuma,honda,hiro}@ertl.jp, watanabe@coi.nagoya-u.ac.jp DSMS DSMS 32% 46% RTOS Priority Inversion Time Reduction by Operator-Level Commit of DSMS Satoshi