Page 1
Renewable Energy-Aware Data Centre Operations for Smart
Cities – the DC4Cities Approach
SMARTGREENS 2015
SONJA KLINGERTUNIVERSITY OF MANNHEIM
D C 4 C I T I E S g r o u p
Follow us! @ D C 4 C I T I E S
Page 2
General Approach
SMARTGREENS 2015
Data Centres in the CityLack of locally produced renewable energy
due to space limitations. -> minimize energy consumption and adhere to constraints of a higher directive – the EMA-SC
Page 3
High-Level Architecture
SMARTGREENS 2015
Page 4
Coordination between SC and DC
SMARTGREENS 2015
A new authority: The energy management authority of the smart city (EMA-SC)
The EMA-SC sets objectives to which the data centres have to adhere to
These are taken into account for calculating an ideal power budget in the DC
In case the DC cannot comply with the objectives an escalation to the EMA is triggered
Page 5
Energy Adaptation within a DC
SMARTGREENS 2015
Multi-level API for IaaS, PaaS and SaaS
Page 6
Incentives and Monitoring
SMARTGREENS 2015
Smart City as mediator between Energy System and DCs
RenEnergy Contract between EMA-SC and DC DCAdapt metric: Deviation between Ideal Power Plan
and realized power profile RenPercent metric: The share of renewable energy
consumed by the DCGreenSLAs: Contracts between DCs and it‘s
costumers allowing more flexibility and can contain metrics describing the guarenteed eco-
efficiency of the service
Page 7
The DC4Cities Architecture
SMARTGREENS 2015
1. DC4cities process controller retrieves the next 24 hours energy forecasts for each EP of the DC through the ERDS handler
2. The Max/Ideal power plan is computed3. The power plan is split into different plans, one for each service hosted by the DC
4. Multiple splitting policies can be configured to better tailor the system to the DC business needs
5. The controller will request EASC to create specific power budgets for the next 24 hours for each service
6. The Option plan collector will receive a set of possible alternatives by each EASC
7. All Option plans will be consolidated and globally optimized to achieve the best usage of renewable energy source
8. If a good solution is found, the EASCs are informed which option plan to enact. Else, an escalation process is triggered [8x]
9. EASC will use automation tools to control the SW/HW resources of the service in line with the received plan (Working Mode).
10. Finally the controller will share the DC power plan with the energy provider, to enable some form of demand/response cooperation
Page 8
DC4Cities - Trials
SMARTGREENS 2015
CPU Intensive video conversion task
Generation of Reports for local health system
Test Lab for a web E-learning platform (worldwide)
Page 9
Results – HP and Trento
SMARTGREENS 2015
Batch jobs: Producing 4320 reports per dayPercentage of Renewable Energy in the Italian
Grid varies between 29,21% and 49,18% (avg. 37,16)
Uniform workload distribution over 24 hours Workload concentrated at grid max RenPerc
37,16% 42,20%
Page 10
Results –HP and Trento (cont.)
SMARTGREENS 2015
When adding 8 local solar panels (max 250Wh) to the previous setting, the RenPercent rises to 79,41%
Local Solar Energy Production
Page 11
QUESTIONS?
SMARTGREENS 2015
Thank you!
K L I N G E R T @ I N F O R M AT I K . U N I -M A N N H E I M . D E
W W W. D C 4 C I T I E S . E U
D C 4 C I T I E S g r o u p
Follow us! @ D C 4 C I T I E S
Contact us!