veyepar: Elfen Scheduling: Fine-Grain Principled Borrowing from Latency-Critical Workloads using SMT

Hi user

Admin Login:

Name: Elfen Scheduling: Fine-Grain Principled Borrowing from Latency-Critical Workloads using SMT
--client lca --show lca2018 --room medium 13586 --force
Prev: 14 Unions: The way to hack society's operating system
Next: 14 You Should Speak

show more...

Author(s):	Xi Yang
Location	Medium Lecture Theatre CB11.00.401
Date	jan Fri 26	Days Raw Files
Start	11:40	First Raw Start	11:22
Duration	0:45:0	Offset	0:17:05
End	12:25	Last Raw End	12:52
Chapters	00:00 0:07:37 0:37:37
Total cuts_time	40 min.

http://lca2018.linux.org.au/schedule/presentation/56/
raw-playlist raw-mp4-playlist encoded-files-playlist host archive public tweet mp4 svg png

final audio

assets release.pdf Elfen_Scheduling_FineGrain_Principled_Borrowing_from_LatencyCritical_Workloads_using_SMT.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:
Locked:	clear this to unlock
Locked by:	user/process that locked.
Start:	initially scheduled time from master, adjusted to match reality
Duration:	length in hh:mm:ss
Name:	Video Title (shows in video search results)
Emails:	email(s) of the presenter(s)
Released:	has someone authorised pubication
Normalise:
Channelcopy:	m=mono, 01=copy left to right, 10=right to left, 00=ignore.
Thumbnail:	filename.png
Description:	Web services from search to games to stock trading impose strict Service Level Objectives (SLOs) on tail latency. Meeting these objectives is challenging because the computational demand of each request is highly variable and load is bursty. Consequently, many servers run at low utilization (10 to 45%); turn off simultaneous multithread- ing (SMT); and execute only a single service — wasting hardware, energy, and money. Although co-running batch jobs with latency critical requests to utilize multiple SMT hardware contexts (lanes) is appealing, unmitigated sharing of core resources induces non-linear effects on tail latency and SLO violations. We introduce principled borrowing to control SMT hardware execution in which batch threads borrow core resources. A batch thread executes in a reserved batch SMT lane when no latency-critical thread is executing in the partner request lane. We instrument batch threads to quickly detect execution in the request lane, step out of the way, and promptly return the borrowed resources. We introduce the nanonap system call to stop the batch thread’s execution without yielding its lane to the OS scheduler, ensuring that requests have exclusive use of the core’s resources. We evaluate our approach for colocating batch workloads with latency-critical requests using the Apache Lucene search engine. A conservative policy that executes batch threads only when request lane is idle improves utilization between 90% and 25% on one core depending on load, without compromising request SLOs. Our approach is straightforward, robust, and unobtrusive, opening the way to substantially improved resource utilization in datacenters running latency-critical workloads. markdown
Comment:	production notes

recording sheet

recording sheet

2018-01-26/11_22_55.ts Apply: 11:22:55 - 11:45:17 ( 00:22:22 ) S: 11:22:55 - E: 11:52:54 D: 00:29:59 (End: 1342.0) show more...

vlc ~/Videos/veyepar/lca/lca2018/dv/medium/2018-01-26/11_22_55.ts :start-time=00.0 --audio-desync=0

	Raw File	Cut List
	11:22:55		seconds: 0.0 Wall: 11:22:55
Duration	00:29:59
	11:52:54		seconds: 1342.0 Wall: 11:45:17
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2018-01-26/11_22_55.ts Apply: 11:45:17 - 11:52:54 ( 00:07:37 ) S: 11:22:55 - E: 11:52:54 D: 00:29:59 (Start: 1342.0) show more...

vlc ~/Videos/veyepar/lca/lca2018/dv/medium/2018-01-26/11_22_55.ts :start-time=01342.0 --audio-desync=0

	Raw File	Cut List
	11:22:55		seconds: 1342.0 Wall: 11:45:17
Duration	00:29:59
	11:52:54		seconds: 0.0 Wall: 11:22:55
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2018-01-26/11_52_55.ts Apply: 11:52:55 - 12:22:55 ( 00:30:00 ) S: 11:52:55 - E: 12:22:55 D: 00:30:00 show more...

vlc ~/Videos/veyepar/lca/lca2018/dv/medium/2018-01-26/11_52_55.ts :start-time=00.0 --audio-desync=0

	Raw File	Cut List
	11:52:55		seconds: 0.0 Wall: 11:52:55
Duration	00:30:00
	12:22:55		seconds: 0.0 Wall: 11:52:55
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2018-01-26/12_22_55.ts Apply: 12:22:55 - 12:25:23 ( 00:02:28 ) S: 12:22:55 - E: 12:52:54 D: 00:29:59 (End: 148.0) show more...

vlc ~/Videos/veyepar/lca/lca2018/dv/medium/2018-01-26/12_22_55.ts :start-time=00.0 --audio-desync=0

	Raw File	Cut List
	12:22:55		seconds: 0.0 Wall: 12:22:55
Duration	00:29:59
	12:52:54		seconds: 148.0 Wall: 12:25:23
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2018-01-26/12_22_55.ts Apply: 12:25:23 - 12:52:54 ( 00:27:31 ) S: 12:22:55 - E: 12:52:54 D: 00:29:59 (Start: 148.0) show more...

vlc ~/Videos/veyepar/lca/lca2018/dv/medium/2018-01-26/12_22_55.ts :start-time=0148.0 --audio-desync=0

	Raw File	Cut List
	12:22:55		seconds: 148.0 Wall: 12:25:23
Duration	00:29:59
	12:52:54		seconds: 0.0 Wall: 12:22:55
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

Rf filename:	root is .../show/dv/location/, example: 2013-03-13/13:13:30.dv
Sequence:
get this:	check and save to add this

2018-01-26/11_22_55.ts
audio viz

audio viz

2018-01-26/11_52_55.ts
audio viz

audio viz

2018-01-26/12_22_55.ts
audio viz

audio viz

Veyepar Video Eyeball Processor and Review