SLIDE 1
A data model for supplying a Data Center with several energy sources
Inès de Courchelle IRIT/LAAS idecourc@irit.fr
18 juillet 2016
Tom Guérout LAAS-CNRS tguerout@laas.fr Yann Labit LAAS-CNRS ylabit@laas.fr Thierry Monteil LAAS-CNRS monteil@laas.fr
SLIDE 2 2 26
Plan
- I. Introduction
- II. RenewSim
- III. Modeling
- IV. Results
- V. Conclusion
SLIDE 3
Introduction RenewSim Modeling Results Conclusion Introduction
3 26
CONTEXT
◆ Smart cities ◆ Home automation ◆ Smart building ◆ Smart Grid ◆ ... ◆ Use renewable energy sources ◆ Reducing the CO2 foot print ◆ Smart Building ◆ Supply a Data Center ◆ The running cost is more and more expensive ◆ Actual Data Center capacities ◆ 40 000 servers ◆ 500 000 services (VM) ◆ > 1 million
SLIDE 4
Introduction RenewSim Modeling Results Conclusion Introduction
4 26
Data Center Consumption
2000 70 Twh in the world 2007 330 Twh, 2% World C02 production 2011 France is the 6th electricity consumer in the world 2020 104 billion kW in European area 2008 56 billion kW in European area
How to supply Data Center ? Economically ? Environmentally ?
SLIDE 5
Introduction RenewSim Modeling Results Conclusion Introduction
5 26
Challenge & Problematic
◆ Use several energy sources ◆ Green energy (solar panels, batteries, etc.) ◆ Brown energy (suppliers, etc.)
The problematic is how to characterize the heterogeneous electricity sources to make a decision to reduce electricity bills ?
◆ Control ◆ Energy Sources ◆ Energy Destinations
SLIDE 6
6 26
Introduction RenewSim Modeling Results Conclusion Introduction
Context
Overview
◆ What is it ? ◆ Green IT Simulator ◆ Co-simulation tool ◆ Based on a micro-grid architecture ◆ Why ? ◆ Use several energy sources ◆ Managing the power flow in a clever way ◆ How is it composed ? ◆ Each element is plugged to the bus ◆ Converter module ◆ Control system
SLIDE 7
7 26
Introduction RenewSim Modeling Results Conclusion Introduction
Control module
◆ Find the best trade-off (electricity vs IT) ◆ Manage the power flow into the smart grid ◆ Task allocation characterize by ◆ CPU & Memory capacities ◆ Energy demand and available servers
RenewSim aims at reducing the energy cost by studying the power flow and the Data Center demand in a decision block.
SLIDE 8
Introduction RenewSim Modeling Results Conclusion Introduction
8 26
The Modeling
◆ Manage the different degrees of freedom ◆ Specify the intrinsic behavior of each block
SLIDE 9
Introduction RenewSim Modeling Results Conclusion Introduction
9 26
Context
Freedom degrees 1/2
◆ No degrees ◆ The power asked by the Data Center does not let us a lot freedom degrees ◆ Input of the system asked by the user and application ◆ Solar production does not depend on the system ◆ weather and time of the day ◆ Freedom degrees ◆ The Suppliers ◆ The Batteries
SLIDE 10
Introduction RenewSim Modeling Results Conclusion Introduction
10 26
Freedom degrees 2/2
◆ Batteries ◆ Store energy from solar panels or suppliers ◆ Unload energy from batteries ◆ The suppliers ◆ Price depends of time (night and day) ◆ Store energy into the batteries or sell it
SLIDE 11
Introduction RenewSim Modeling Results Conclusion Introduction
11 26
Element characterization
Source (St) Backup (Bt) Destination (Dt) Overflow (Ot) Gives energy to destinations Receives energy from the sources Stores energy which can not be accepted by the destinations Unloads energy which can be feed the destinations
SLIDE 12
Introduction RenewSim Modeling Results Conclusion Introduction
12 26
Example (1/3)
◆ The modeling is presented as a matrix with 4 columns ◆ Each line is a component of our system
Source Destination Backup Overflow Solar Panels 1 Data Center 1 Battery 1 1 Supplier 2 2
SLIDE 13
Introduction RenewSim Modeling Results Conclusion Introduction
13 26
Example (2/3)
1 5 2 5 5 5 Source Destination Backup 1 Overfull 1 Backup 2 Overfull 2
SLIDE 14
Introduction RenewSim Modeling Results Conclusion Introduction
14 26
Example (3/3)
1 5 2 5 Source Destination Backup 1 Overfull 1 Backup 2 Overfull 2 5
SLIDE 15
Introduction RenewSim Modeling Results Conclusion Introduction
15 26
Algorithms
SLIDE 16
16 26
Introduction RenewSim Modeling Results Conclusion Introduction
Study area (1/2)
◆ Converter ◆ CM1: 15% ◆ CM2: 17% ◆ CM3: 13% ◆ CM4: 14% ◆ ADREAM ◆ Data solar panels: 48h ◆ 1648 measures ◆ Inputs
SLIDE 17
17 26
Introduction RenewSim Modeling Results Conclusion Introduction
Inputs (1/2)
SLIDE 18
18 26
Introduction RenewSim Modeling Results Conclusion Introduction
Inputs (2/2)
SLIDE 19
19 26
Introduction RenewSim Modeling Results Conclusion Introduction
Study area (2/2)
Load unload Capacity 10 000 joules
Battery Supplier
Sale Purchase d
SLIDE 20
20 26
Introduction RenewSim Modeling Results Conclusion Introduction
Experimentation 1
Source Destination Backup Overflow Solar Panels 1 Data Center 1 Battery 1 1 Supplier 2 2
SLIDE 21
21 26
Introduction RenewSim Modeling Results Conclusion Introduction
Battery and supplier
SLIDE 22
22 26
Introduction RenewSim Modeling Results Conclusion Introduction
Experimentation 2
Day Source Destination Backup Overflow Solar Panels 1 Data Center 1 Battery 1 1 1 Supplier 2 2 Night Source Destination Backup Overflow Solar Panels 1 Data Center 1 Battery 1 Supplier 1
SLIDE 23
23 26
Introduction RenewSim Modeling Results Conclusion Introduction
Battery and supplier
SLIDE 24
24 26
Introduction RenewSim Modeling Results Conclusion Introduction
Conclusion
◆ Renewsim allows: ◆ Design a Green modular architecture ◆ Simulate various kind of inputs and storage models ◆ Manage the power redirection through labeled and prioritized energy
sources
◆ Analyze the impact of the IT feed on energy sources
SLIDE 25
25 26
Introduction RenewSim Modeling Results Conclusion Introduction
Perspectives
◆ The expansion prospects consist in: ◆ Heuristics, algorithms of learning machine ◆ Prediction of production & usage ◆ Tasks priority ◆ QoS management ◆ The purpose of RenewSim is to be a co-simulation tool allowing communication
simulators dedicated to each component.
SLIDE 26
Any Questions ?
Inès de Courchelle IRIT/LAAS idecourc@irit.fr Tom Guerout LAAS-CNRS tguerout@laas.fr Yann Labit LAAS-CNRS ylabit@laas.fr Thierry Monteil LAAS-CNRS monteil@laas.fr
SLIDE 27
27 26
SLIDE 28
28 26
SLIDE 29
29 26
Introduction RenewSim Modeling Results Conclusion Introduction Introduction Related Work Micro-grid Modeling Results Introduction
Context
Example
5 1 5 5 1 5