- 追加された行はこのように表示されます。
- 削除された行は
このように表示されます。
!ASPLOS2日目
昨日は昼間いい感じに目が覚めててよかっと思ったものの，
夕方激しい睡魔におそわれ睡眠...からの深夜に起床．
ワークショップの一部がハンズオン的な感じだったこともあって
割と一日有意義に起きてられてよかった．
!Unlocking the Power of Edge Computing
:: Towards Special-purpose Edge Computing
What are architectural and research challenges for realizing specialized edge computing?
* two-tier: cloud <-> IoT device
* three-tier: cloud <-> Edge node <-> IoT device, cloud <-> cloudlet <-> IoT device,
** cloudlet: edge data center, distributed edge cloud
* two-tier specialized edge: edgenode(+ VPU/TPU) <-> IoT devices
** specialized edge nodes
** accelerate specific workloads
** "server-class" performance
** Little/no cloud reliance
* two-tier specialized edge variants: edgenode(+ VPU/TPU) <-> IoT devices(with accelerator)
** tensorflow board, GAP8 IoT processor
* Research challenge
** specialized edge can only run a single class of application
*** lower hardware reuse across application classes(no multi-tenancy)
*** multiple specialized hardware configurations needed to support different application classes
***- increases hardware costs and management complexity
* Cloud vs. Edge economics
** cloud: greater multiplexing benefits
** edge: smaller number of servers per site - lower smoothing: reduces multiplexing benefits
** lower ecnomy of scale for edge clouds
* Hardware Heterogeneity
* split application processing
** application needs to be distributed across tiers - what function to put where?
* Challenges
** greater hardware complexity
** greater application complexity
* Macroprogramming
** origins in sensor networks (circa 2005)
** specify aggregate system behavior rather than device behavior
*** hides hardware diversity from programmers
*** write onece, run anyware
* cf. http://lass.cs.umass.edu/
:: Live Video Analytics - the “killer app” for edge computing!
Video analytis towards vision zero
** cf. https://www.microsoft.com/en-us/research/publication/video-analytics-towards-vision-zero/
** cf. https://www.psrc.org/sites/default/files/peer1707-pres-videoanalytics.pdf
* Democratize ideo analytics
** real-time, low-cost, accurate
* video analytics at scale with approximation, NSI17, SIGCOMM18, SEC18 OSDI18
** video pipeline optimizer sigcomm18
** resource manager nsdi17
** edge/cloud executer sec18
** camera manager ipsn18
** video event store osdi18
* low-cost ingestion: chaper CNN
** cheap CNNs are less accurate
** cheap CNNs can achieve high recall with small top-K results
** -> solution: top-k approximate index
* low-latency query: redundancy elimination
* cf. https://www.microsoft.com/en-us/research/project/live-video-analytics/
* cf. https://github.com/antriv/MLADS_FALL_GAN_2017/tree/master/ppt
::Edge-to-cloud computing infrastructure inspired by the emerging needs of Telco applications
cf. https://www.lfedge.org/projects/akraino/
cf. https://www.o-ran.org/
::Edge computing in the extreme and its applications
cf. https://paradrop.org/
cf. https://paradrop.readthedocs.io/en/latest/index.html
3rd party apps/service drop into your home WiFi router on-demand
!Programming Quantum Computers: A Primer with IBM Q and D-Wave Exercises
https://arcb.csc.ncsu.edu/~mueller/qc/qc-tut/
::IBM Q -- Quantum Gate Programming
* Quantum Algorithm Strategies
** create supoerposition of states
** apply transforms that amplify desirable values and diminish unwanted values
::D-Wave
::Quirk
https://algassert.com/2016/05/22/quirk.html
https://github.com/Strilanc/Quirk/tree/master/src/base
* [toffoli|https://algassert.com/quirk#circuit=%7B%22cols%22%3A%5B%5B%22H%22%2C%22H%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%2C%22Bloch%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C%22Bloch%22%2C%22Bloch%22%5D%2C%5B1%2C1%2C1%2C%22H%22%5D%2C%5B1%2C1%2C1%2C%22Bloch%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C%22Bloch%22%2C1%2C%22Bloch%22%5D%2C%5B1%2C1%2C1%2C%22Z%5E-%C2%BC%22%5D%2C%5B1%2C1%2C1%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C1%2C%22X%22%5D%2C%5B%22Bloch%22%2C1%2C1%2C%22Bloch%22%5D%2C%5B1%2C1%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C1%2C1%2C%22Bloch%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C%22Bloch%22%2C1%2C%22Bloch%22%5D%2C%5B1%2C1%2C1%2C%22Z%5E-%C2%BC%22%5D%2C%5B1%2C1%2C1%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C1%2C%22X%22%5D%2C%5B%22Bloch%22%2C1%2C1%2C%22Bloch%22%5D%2C%5B1%2C%22Z%5E%C2%BC%22%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%2C1%2C%22Bloch%22%5D%2C%5B1%2C1%2C1%2C%22H%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22Z%5E%C2%BC%22%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%5D%7D]
* [swap|https://algassert.com/quirk#circuit=%7B%22cols%22%3A%5B%5B%22X%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22X%22%2C%22%E2%80%A2%22%5D%2C%5B%22Bloch%22%2C%22Bloch%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%5D%7D]
* [Grover Alg.|https://algassert.com/quirk#circuit=%7B%22cols%22%3A%5B%5B%22H%22%2C%22H%22%2C%22H%22%2C%22H%22%5D%2C%5B%22Sample4%22%5D%2C%5B1%2C%22X%22%2C%22X%22%2C%22X%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C%22X%22%2C%22X%22%2C%22X%22%5D%2C%5B%22H%22%2C%22H%22%2C%22H%22%2C%22H%22%5D%2C%5B%22X%22%2C%22X%22%2C%22X%22%2C%22X%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C1%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E%C2%BC%22%5D%2C%5B1%2C%22%E2%80%A2%22%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E-%C2%BC%22%5D%2C%5B%22%E2%80%A2%22%2C1%2C%22X%22%5D%2C%5B1%2C1%2C%22%E2%80%A2%22%2C%22Z%5E%C2%BC%22%5D%2C%5B%22X%22%2C%22X%22%2C%22X%22%2C%22X%22%5D%2C%5B%22H%22%2C%22H%22%2C%22H%22%2C%22H%22%5D%5D%7D]