Data centers & energy: Did we get it backwards? Adam Wierman, Caltech
The typical story about energy & data centers:
The typical story about energy & data centers: Sustainable data centers Remember: The cloud is (often) more efficient than the alternative.
The typical story about energy & data centers: But maybe we got it backwards?
Renewable energy is coming! but incorporation into the grid isn t easy
Today s grid Generation Load Key Constraint: Generation = Load (at all times) low uncertainty
Today s grid Generation Load Key Constraint: Generation = Load (at all times) controllable (via markets) low uncertainty
Tomorrow s grid Key Constraint: Generation = Load (at all times) less controllable high uncertainty low uncertainty
1) Huge price variability, leading to generators opting out of markets! 2) More conventional reserves needed, countering sustainability gains! Key Constraint: Generation = Load (at all times) less controllable high uncertainty low uncertainty
1) Huge price variability, leading to generators opting out of markets! 2) More conventional reserves needed, countering sustainability gains! Key Constraint: Generation = Load (at all times) less controllable high uncertainty low uncertainty
Grid needs huge growth in demand response (or storage) Data centers are a promising option they are large loads usage is growing quickly highly automated they have significant flexibility 10-15% growth/year 500 kw-100 MW each
Building management 5% in 2 min / 10% in 20min [LLNL] e.g. cooling, lightin cost g, flexibility Data centers are a promising option they are large loads usage is growing quickly highly automated they have significant flexibility 10+ years of research into energy-efficient data centers
Building management 5% in 2 min / 10% in 20min [LBNL] e.g. cooling, lighting, cost flexibility Data centers are a promising option they are large loads usage is growing quickly highly automated they have significant flexibility 10+ years of research into energy-efficient data centers
Workload management 10-30+% in 10-60min [LBNL,HP] e.g. demand shaping, geographical load balancing, quality degradation, cost flexibility Data centers are a promising option they are large loads usage is growing quickly highly automated they have significant flexibility 10+ years of research into energy-efficient data centers
Microgrid management 10-100% in 5-30min e.g., Battery management, local PV, Backup generation cost flexibility Data centers are a promising option they are large loads usage is growing quickly highly automated they have significant flexibility 10+ years of research into energy-efficient data centers
A new story about energy & data centers: Data centers are valuable resources for making the grid sustainable =
What is the potential of data center demand response?
A case study: PV data center interactive workload batch workload Optimally placed, fast charging rate storage PUE
A case study:
1 MWh if geographical load balancing is used! A case study: $1-5 million cost!
Where are we today? Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergency DR Data centers rarely participate and if they do it is highly inefficient
Where are we today? coincident peak warnings peak Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergency DR - Risky to participate - Few opportunities for utility to extract response For more see [Liu et al 2013]
Where are we today? Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergency DR
How can we do better? Engineering: + [Camacho Economics: Algorithm design for data center participation et al 2014], [Chen et al 2013, 2014], [Ghamkhari et al 2012, 2014], [Aikema et al 2012, 2013], [Irwin et al 2011], [Urgaonkar et al 2013, 2014], [Li et al 2012, 2013], [Liu et al 2013, 2014] New market designs [Liu et al 2014], [Chen et al 2014],[Ghamkhari et al 2013], [Li et al 2013], [Wang et al 2014], [Ren et al 2015], [Wierman et al 2014], [Zhang et al 2015]
How can we do better? Engineering: + Economics: Performance is priority #1 Algorithm design for data center participation Highly regulated, change is difficult New market designs but, adoption represents a huge challenge
A starting point: Colocated (multi-tenant) data centers
A starting point: Colocated (multi-tenant) data centers Hyper-scale (e.g. google): 7.8% Enterprise: 53% Colocation: 37% of total data center industry electricity usage 29
Why colocated data centers? Building operation is separated from computing priorities but the data center would like to participate in demand response! Set up incentives for tenants + On-site generation provides backup! + Market power isn t an issue! + No regulation can do whatever they want! + Tenants are heterogeneous in workloads! Lot s of work to be done
The typical story about energy & data centers: But maybe we got it backwards? Key points: 1) We need to move beyond a myopic focus on a data centers to consider a system-wide view of sustainability. 2) It is important to consider more than google-like hyper-scale data centers.
Data centers & energy: Did we get it backwards? Adam Wierman, Caltech