ee5900 advanced embedded system for smart infrastructure
DESCRIPTION
EE5900 Advanced Embedded System For Smart Infrastructure. Computationally Efficient Smart Home Scheduling. 3. 1. 2. 4. 5. Case Study. Conclusion. Smart Home. Cloud Computing. Algorithm. Outline. 2. Smart Home. Power Line. Communication Line. 3. End. Start. Dish washer. 13:00. - PowerPoint PPT PresentationTRANSCRIPT
EE5900 Advanced Embedded System For Smart Infrastructure
Computationally Efficient Smart Home Scheduling
Smart Home1
Cloud Computing2
Case Study4
Algorithm3
Outline
2
Conclusion5
Smart Home
3
Power Line
Communication Line
Landry machineDish washer
PHEVAC
Start End
……
13:00 18:0009:00 18:00
08:0018:0017:00 N/A
4
Home Appliance (HA) in Smart Home
5
Non-schedulable HA Restrictive-schedulable HA
Full-schedulable HA
Multiple Power Levels
6
350 W
Power level
500 W820 W
1350 W
http://www.supplyairconditioner.com/1-4-9-split-wall-mounted-air-conditioner.html
Multiple Working Stages
7
Working cycles
Prewash
Washing
Rinsing
Spinning
Assume all stages have same working frequency for simplicity Partition the whole task to multiple subtasks with precedence constraints
Drying
Plug-in Hybrid Electric Vehicles (PHEV)
8
Powered by an Electric Motor and Engine
• Internal combustion engine uses alternative or conventional fuel
• Battery charged by outside electric power source, engine, or regenerative breaking
• During urban driving, most power comes from stored electricity. Long trips require the engine
Contemporary Hybrids
9
Toyota PriusToyota Camry Toyota Highlander Honda Insight
Lexus RX400h Lexus GS450h
Honda Civic Honda Accord
Saturn Vue Chevy Silverado
Ford Escape
Charging of PHEV
10
Level 1: 120 V, alternating current (AC) plug; dedicated circuit Level 2: 240 V, AC plug and uses the same connector on the vehicle as Level 1 Level 3: In development; faster AC charging
Existing Products of Battery
Accord PHEV 120-volt: less than 3 hours 240-volt: one hour
Toyata PHEV 120-volt: less than 3 hours 240-volt: 1.5 hours
Quick charge to 80% needs 30 minutes.
11
Dynamic Pricing from Utility Company
12
https://rrtp.comed.com/live-prices/?date=20130404
Dynamic Voltage and Frequency Scaling (DVFS)
13
10 cents/kwh 5 cents / kwh
5 kwh
10 kwh
Power Powerr
Time Time1 2 1 2 3
(a) (b)
10 cents/kwh 5 cents / kwh
cost = 10 kwh * 10 cents/kwh = 100 cents cost = 5 kwh * 10 cents/kwh + 5 kwh * 5 cents/kwh = 75 cents
Smart Home Scheduling (SHS)
14
Given n home appliances, to schedule them for monetary cost minimization satisfying the total energy constraint and deadline constraints
Demand Side Management– when to launch a home appliance– at what frequency
– The variable frequency drive (DVFS) is to control the rotational speed of an alternating current (AC) electric motor through controlling the frequency of the electrical power supplied to the motor
– for how long
Benefit of Smart Home
15
– Reduce monetary expense
– Reduce peak load
Smart Home Scheduling (SHS)
16
Home appliance level
User level
Community level
Smart Home Scheduling (SHS)
Home appliance level
User level
Community level
17
Single Home appliance Scheduling
18
Non-schedulable HA
Consider the non-schedulable home appliance as fix energy consumption
Single Home appliance Scheduling
19
For restrictive-schedulable home appliance, set start time to be earlier than the user’s requirement.For example, in summer, user wants to come back to home at 5pm. The AC should be on before 5pm.
Restrictive-schedulable HA
Single Home appliance Scheduling
20
For full-schedulable home appliance, one needs to schedule when to launch a home appliance at what frequency considering DVFS for how long to minimize monetary cost satisfying that the total energy is consumed.
Full-schedulable HA
Home Appliance Definition
21
Ts: Start time Te: End time Pi: Power level E: Total required energy : Unit price of time slot t
Dynamic Programming
22
Given a home appliance, one processes time slot one by one for all possibilities until the last time slot and choose the best solution 𝑇 𝑠 𝑇 𝑠+1 𝑇 𝑠+2 𝑇 𝑒−1 𝑇 𝑒
0
𝑃1
𝑃2
0
𝑃1
𝑃2
0
𝑃1
𝑃2
Choose the solution with total energy equal to E and minimal monetary cost
Characterizing
23
For a solution in time slot i, energy consumption e and cost c uniquely characterize its state
Time slot i Time slot i+1(ei, ci) (ei+1, ci+1)
Pruning
24
For one time interval, (e1, c1) will dominate solution (e2, c2), if e1>= e2 and c1<= c2
Time slot i(15, 20)
(15, 25)
(11, 22)
Algorithmic Flow of Dynamic Programming
25
Calculate all possible (e, c)
Prune all dominated (e, c)
Choose the result (e, c) which e = E and c is minimal
Schedule
Start time t = Ts
Yes
Next time slott = t + 1
End time t = TeNo
No Schedulee < E
Dynamic Programming based Appliance Optimization
26
(1,2)
(2,4)
(3,6)
(1,1)
(2,2)
(3,3)
0 t1 t2
(6, 9) (5, 8)(4, 7)
(5, 7) (4, 6)(3, 5)
(4, 5) (3, 4)(2, 3)
(0,0) (0,0)
(3, 3) (2, 2)(1, 1)
– # of distinct power levels = k– # time slots = m)( 2kmORuntime :
Price
Time
Dynamic Programming returns optimal solution
𝛼1=2 𝛼2=1
Power level: {1, 2, 3}
Smart Home Scheduling (SHS)
Home appliance level
User level
Community level
27
Scheduling Among Multiple Appliances for One User
28
Determine Scheduling Appliances Order
Schedule Current Task
Update Upper Bound of Each Time Interval
An appliance
Schedule
Appliances
Not all the appliance(s) processed
All appliance process
Smart Home Scheduling (SHS)
Home appliance level
User level
Community level
29
0 2 4 6 8 10 12 14 16 18 20 22 2402468
101214
Price CurveGame Approach
User 1 User 2 User m.............
A game approach is deployed where each customer acts as a player.
30
0 2 4 6 8 10 12 14 16 18 20 22 2402468
101214
Price Curve
Game Theory
31
For every player in a game, there is a set of strategies and a payoff function, which is the profit of the player.
Each player choose actions from the set of strategies in order to maximize its payoff.
When no player can increase its payoff without changing the actions of others, Nash Equilibrium is reached.
Game Formulation in Community Level
32
Players: All the customers in the community
Payoff:
Strategy: Choose power levels and launch time to maximize payoff while the constraint conditions can be satisfied
Algorithmic Flow in Community Level
33
Each user schedules their own appliances separately
All users share information with each other
Each user reschedules their own appliances separately
Schedule
Equilibrium
Yes
No
Multiple Customer Scheduling
34
u1 u2 u3
r1 r2 r3
Communication
FPGA First iteration
Communication
FPGA FPGA
u1 u2 u3
Second iterationFPGA FPGA FPGA
……Schedule
• Low frequency• High cost• Hard to maintain
Equilibrium
……
Cloud Computing
35
In Cloud Computing, a new class of network based computing takes place over the Internet
It is a collection/group of integrated and networked hardware, software and Internet infrastructure
Why Cloud Computing
36
Advantages– Low cost– High availability, flexibility, elasticity
– You can increase or decrease capacity within minutes, not hours or days;
– You can commission one, hundreds or even thousands of server instances simultaneously.
– Your application can automatically scale itself up and down depending on its needs.
– Free of maintenance– Security
Service modelsSoftware as a
Service (SaaS)Platform as a
Service (PaaS)Infrastructure as a
Service (IaaS)
Google App Engine
SalesForce CRMLotusLive
37
Cloud Taxonomy
38
Some Commercial Cloud Offerings
39
Amazon EC2
40
Amazon EC2 is one large complex web service. EC2 provided an API for instantiating computing
instances with any of the operating systems supported.
It can facilitate computations through Amazon Machine Images (AMIs) for various other models.
Google App Engine
41
This is more a web interface for a development environment that offers a one stop facility for design, development and deployment Java and Python-based applications in Java and Python.
Google offers the same reliability, availability and scalability at par with Google’s own applications
Interface is software programming based
Windows Azure
42
Enterprise-level on-demand capacity builder Fabric of cycles and storage available on-request for a cost You have to use Azure API to work with the infrastructure
offered by Microsoft
In Home vs. Cloud Computing Scheduling
43
Cost– High performance FPGA vs. Low performance FPGA + Cloud
– Low performance FPGA vs. Low performance FPGA + Cloud Upgrade
– Upgrade FPGA vs. Cloud service Maintenance
– Broken FPGA– Cloud is free of maintenance
Runtime– In Home vs. Cloud Computing
Estimation of Computation Time of Low Performance FPGA FPGA in smart home: 250 MHz
– 1000 users with 1000 FPGA– Runtime is approximately 10 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (10+0.04)*100 = 1004 sec = 16.73 min
Since the pricing policy is updated each 15 minutes by most utilities, 16.73 minutes are unacceptable.
Why not using some quite high performance machines in each home?
44
Cloud Based Distributed Algorithm
45
u1 u2 u3
r1 r2 r3
Communication
FPGA
First iteration
Communication
FPGA FPGA
……Schedule Equilibrium
……
FPGA FPGA FPGA
r1 r2 r3
u1 u2 u3
Cloud
Monetary Cost Aware Scheduling Problem
There are different types of machines in cloud with different monetary cost, frequencies and storage
One is required to schedule those users’ tasks to appropriate machines to minimize the monetary cost of the distributed algorithm satisfying the timing constraints
46
Runtime (s) u1 u2 u3 u4
FPGA 12 14 10 15
2 GHz 1.5 1.75 1.25 1.88
3 GHz 1 1.17 0.83 1.25
An example I
47
FPGA: 250 MHz CPU in cloud: 2 GHz with $0.02/hour, 3 GHz with $0.06/hour Timing constraints Tc = 5
The monetary cost C = 1.25 / 3600 * 0.02 + (1+1.17+1.25) / 3600 * 0.06 = $6.39 * 10 -5.
If one schedules tasks of user 3 to CPU with 2 GHz and schedules tasks of user 1, 2 and 4 to CPU with 3 GHz, then
The runtime T = max{1.25, 1+1.17+1.25} = 3.42 < Tc.
An example II
48
FPGA: 250 MHz CPU in cloud: 2 GHz with $0.02/hour, 3 GHz with $0.06/hour Timing constraints Tc = 5
u1 u2 u3 u4
Runtime (s) 12 14 10 15
2 GHz (s) 1.5 1.75 1.25 1.88
3 GHz (s) 1 1.17 0.83 1.25
The monetary cost C = (1.5 + 1.75) / 3600 * 0.02 + (0.83 + 1.25) / 3600 * 0.06 = $5.27 * 10 -5.
If one schedules tasks of user 1 and 2 to CPU with 2 GHz and schedules tasks of user 3 and 4 to CPU with 3 GHz, then
The runtime T = max{1.5 + 1.75, 0.83 + 1.25} = 3.25 < Tc.
Problem Formulation
49
Given users in smart home scheduling problems with runtime running in local machine with frequency , types of machines in cloud with frequency and monetary cost , one needs to schedule these users’ tasks to machines such that the total monetary cost is minimized and maximum runtime over all the machines satisfies the timing constraints.
Monetary Cost Problem Formulation
50
Linear Programming With Rounding
51
For each , round the largest to be 1, others to 0
Algorithmic Flow
52
Solve the continuous fashion problem combinatorially
Discretize the continuous
solution
Flag all machine to be available
Assign task fractionally to the available machine with highest
ratio of /𝒄 𝒇Sort all machines increasingly by by
ratio of /𝒄 𝒇 Runtime of machine is reaching TC
Flag the machine to be unavailabe
Yes
No
Combinatorial solving
……
f1 f2 fm-1 fm
TC
……
53
Tc – Timing constraintsfi - Frequency of cloud machines
Discretization
54
f1 f2
TC
12
3
f1 f2
TC
12
3T’
(a) (b)
3
Since we always round fractional scheduled task to machine with smaller ratio , the total monetary cost must be no greater than the optimal solution while the timing constraint may be violated by𝑇 𝐴𝐿𝐺≤𝑇𝐶+max 𝑡 𝑖 ∙(
𝑓 𝑚𝑎𝑥
𝑓 𝑚𝑖𝑛−1)
Theorem
55
There exists an algorithm such that the total monetary cost must be no greater than the solution of continuous problem while the timing constraint may be violated by , running in time
High Level Algorithm
56
The distributed algorithm needs multiple iterations to achieve the equilibrium, thus the scheduling algorithm needs to handle all the iterations repeatedly.
u1 u2 u3
r1 r2 r3
Communication
FPGA
First iteration
Communication
FPGA FPGA
……Schedule Equilibrium
……
FPGA FPGA FPGA
r1 r2 r3
u1 u2 u3
Cloud
FPGA & Amazon EC2
Low performance FPGA in smart home: 250 MHz Computer in cloud:
– 1 core with 1 ECU (approx.. 1.7 GHz, $0.034 per hour)– 1 core with 2 ECU (approx.. 3.5 GHz, $0.068 per hour)– 2 cores with 2 ECU (approx.. 3.5 GHz, $0.136 per hour)– 4 cores with 2 ECU (approx.. 3.5 GHz, $0.271 per hour)
Communication time: 10kb/250kb/s=0.04s
Observing that there are machines with multiple cores, we can schedule multiple tasks to one machine with multiple cores at the same time
57
Comparison for 1000 users
W/o cloud– 1000 users with 1000 FPGA– Runtime is approximately 14 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (10+0.04)*100 = 1004 sec = 16.73 min
W/ cloud of 1 core with 1 ECU – 1000 computers in cloud– Runtime is approximately 2 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (2+0.04)*100 = 3.4 min (4.92X)
58
Comparison for 1000 users
W/ cloud of 1 core with 2 ECU – 1000 computers in cloud– Runtime is approximately 1 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (1+0.04)*100 = 1.7 min (9.84X)
W/ cloud of 4 core with 2 ECU (Parallel in four cores)– 250 computers in cloud– Runtime is approximately 1 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (1+0.04)*100 = 1.73 min (9.67X)
59
Case Study Setup
60
Low performance FPGA in smart home: 250 MHz, $200 High performance FPGA in smart home: 1250 MHz, $2000 Computer in cloud:
– 1 core with 1 ECU (approx.. 1.7 GHz, $61/yr upfront, $0.034/hr)– 1 core with 2 ECU (approx.. 3.5 GHz, $122/yr upfront, $0.068/hr)– 2 cores with 2 ECU (approx.. 3.5 GHz, $243/yr upfront, $0.136/hr)– 4 cores with 2 ECU (approx.. 3.5 GHz, $486/yr upfront, $0.271/hr)
http://www.xilinx.com/support/documentation/data_sheets/ds160.pdfhttp://www.amazon.com/C3-DRK-Digital-Radio-Kit/dp/B001KBPIOQ/ref=sr_1_8?s=pc&ie=UTF8&qid=1365106998&sr=1-8&keywords=fpgahttp://aws.amazon.com/ec2/pricing/
Case Study Setup (Cont.)
61
Home appliances category – Restrictive-schedulable
– Full-schedulable
– Non-schedulable
Frequency level: 20Hz, 40Hz, 60Hz, 80Hz
Start time: 16:00End time: 18:00
Start time: 0:00End time: 23:59
Start time: 9:00
Frequency level: 20Hz, 40Hz, 60Hz, 80Hz
End time: 18:00
Case Study Setup (Cont.)
62
200 to 1000 users in one community Each user could have 10 – 30 home appliance
– 30% of restrictive-schedulable home appliance– 50% of full-schedulable home appliance– 20% of non-schedulable home appliance
An Example – One User
63
HA Start time End time Total energy (kW.h)
Power levels (W)
AC 17:00 20:00 8 {400, 600, 800, 1000,
3000}Washer &
Dryer09:00 18:00 5 1000
Dish Washer 09:00 18:00 3 1000
PHEV 18:00 07:00 12 {1900, 3000, 20k, 240k}
Refrigerator 00:00 23:59 1.2 50
http://www .mpoweruk. com/electr icity_dema nd.htm
64
Total Bill – Monthly
n=20
0n=
400
n=60
0n=
800
n = 10
000
20406080
100120140160180200
Utility Bill W/o SHSUtility Bill w/ Low Performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
65
Runtime
n=20
0n=
400
n=60
0n=
800
n = 10
0002468
1012141618
Runtime of Low per-formance FPGA In Home SHSRuntime of Cloud SHS
Minutes
66
High Performance FPGA
FPGA in smart home: 1250 MHz, $2000
Runtime– 1000 users with 1000 FPGA– Runtime is approximately 2 seconds in one iteration– Communication time: 10kb/250kb/s=0.04s– 100 iterations: (2+0.04)*100 = 204 sec = 3.4 min– No real time issue
67
Total Bill – First Year
n=20
0n=
400
n=60
0n=
800
n = 10
000
5001000150020002500300035004000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
68
Total Bill – Ten Years
n=20
0n=
400
n=60
0n=
800
n = 10
000
5000
10000
15000
20000
25000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
69
Total Bill – Ten Years Cloud computing service cost reduction
n=20
0n=
400
n=60
0n=
800
n = 10
000
5000
10000
15000
20000
25000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
Cloud computing service cost reduction rate: 10%/yr
70
Total Bill – Ten Years FPGA Maintenance
n=20
0n=
400
n=60
0n=
800
n = 10
000
5000
10000
15000
20000
25000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
FPGA maintenance cost: $50/yr
71
Total Bill – Ten Years FPGA Broken
n=20
0n=
400
n=60
0n=
800
n = 10
000
5000
10000
15000
20000
25000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
FPGA broken rate: 2.8%http://homepages.cae.wisc.edu/~aminf/FCCM09%20-%20FPGA%20Design%20Analysis%20of%20the%20Clustering%20Algorithm%20for%20the%20CERN%20Large%20Hadron%20Collider.pdf
72
Total Bill – Ten Years US Dollars Inflation
Inflation rate of US dollars: 2%/yr
http://www.usinflationcalculator.com/inflation/historical-inflation-rates/
n=20
0n=
400
n=60
0n=
800
n = 10
000
5000
10000
15000
20000
25000
30000
Utility Bill W/o SHSUtility Bill w/ High performance FPGA In Home SHSUtility Bill w/ Cloud SHS
Dollars
Conclusion
73
According to case study, our approach by use of cloud can make several times speed up comparing to low performance FPGA based algorithms such that the timing constraints could be satisfied and archive 18.95% monetary cost reduction on average
If high performance FPGA is chosen, user needs to pay 58.3% on average more than bill without SHS in first year of buying FPGA; user will pay higher than cloud based scheme considering cost reduction of cloud computing, maintenance and broken of FPGA in first ten years
Overall, cloud computing is better than both low performance FPGA and high performance FPGA
Further Study
74
Design an algorithm to decide the number of machines in cloud to minimize the reservation cost
More case study will be conducted to generalize my conclusion
Thanks
75