Understanding Power Usage Effectiveness (PUE) & Data Center Infrastructure Management (DCIM) Salim Janbeh Physical Infrastructure Consultant uae-sj@panduit.com PANDUIT
You Can t Manage What You Don t Measure: A Predictive Approach to Energy Efficiency 2
Agenda: 3 1. Data Center Challenges 2. Power Usage Effectiveness Overview 3. Data Center Infrastructure Management Software Requirement
In an environment of constant Change Increasing availability expectations Cloud computing Documentation requirements Uncertain planning for capacity or density Regulatory requirements Energy Efficiency Metrics and reporting Virtualization Infrastructure management Energy and service cost control Dynamic power variation Server consolidation Data center managers need accurate and timely information to make better decisions in real time
BMS Power Management Cooling Control VM Management Server Management CMDB IT Tickets Too much data in siloes systems Facility Management IT Management
BMS Power Management Cooling Control VM Management Server Management CMDB IT Tickets Seamless integration of Key data Facility Management IT Management Data Center Infrastructure Management (DCIM)
DCIM becomes the centerpiece for operational efficiency Facility Management (BMS, Power Management, Cooling control, etc.) DCIM Connectivity management Asset management Capacity Planning Power & environmental Workflow generation Reporting IT Systems, VM Management, Etc. Data collection, meters, sensors, etc.
Agenda: 8 1. Data Center Challenges 2. Power Usage Effectiveness Overview 3. Data Center Infrastructure Management Software Requirement
About the Green Grids 9 A not-for-profit global consortium focused on driving energy efficiency in the computing ecosystem. Developing meaningful and user-centric metrics to help IT and Facilities better manage their computing resources. Developing and promoting standards, measurements, methods, best practices and technologies that support these metrics.
Why Create Metrics? 10 Awareness If you can t measure it, you can t improve it Benchmarking Continuous improvement of operations Comparing across the industry Improvement over time/generation to generation Validate claims Consistency Common vocabulary Industry & Government Collaboration Better Efficiency
Data Center Power Efficiency Metrics: PUE and DCiE PUE= Power Usage Effectiveness = 1 DCiE = Data Center = = Infrastructure Efficiency PUE Total Facility Power IT Equipment Power IT Equipment Power Total Facility Power
PUE Overview 12 PUE provides a way to: Improve a data center s operational efficiency, Compares with similar data centers, Improving the designs and processes over time Reduce infrastructure-energy consumption Target or goal for new data centers. Source: PUE - A Comprehensive Examination of the Metric
PUE Overview 13 Illustration of how PUE would be calculated in a data center Source: PUE - A Comprehensive Examination of the Metric
PUE Overview PUE CALCULATION CONSIDERATIONS 14 Measurement IT equipment energy Total facility energy Measurement interval Level 1 (PUE1) L1 Basic Level 2 (PUE2) L2 Intermediate Level 3 (PUE3) L3 Advanced UPS outputs PDU outputs IT equipment input Utility input Utility input Utility input Monthly Daily Continuous (15 min or less) Source: PUE - A Comprehensive Examination of the Metric
PUE Overview Three PUE measurement levels for a typical data center 15 Source: PUE - A Comprehensive Examination of the Metric
PUE Overview Required and Recommended Measurement Points and Intervals 16 Measurement IT equipment energy Total facility energy Measurement interval Level 1 (PUE1) L1 Basic Level 2 (PUE2) L2 Intermediate Level 3 (PUE3) L3 Advanced Required UPS outputs PDU outputs IT equipment input Required Additional recommended* Required Additional recommended* Utility input Monthly Weekly Utility input UPS inputs/outputs Mechanical inputs Daily Hourly Utility input PDU outputs UPS inputs/outputs Mechanical inputs 15 min 15 min or less *Recommended measurements are in addition to the required measurements. The additional measurement points are recommended to provide further insight into the energy efficiency of the infrastructure. Source: PUE - A Comprehensive Examination of the Metric
PUE Overview Required and Recommended Measurement Points 17 Source: PUE - A Comprehensive Examination of the Metric
PUE Overview Critical Power Path Measurement Points 18 Moving the monitoring location closer to the devices that are consuming the energy enables further isolation of distribution component losses. Dividing the Power path in multiple subzone enable better insight power consumption and and where possible efficiency gains can be made The data center operator must deal with multiple data collection systems which can be streamlined as integrated-measurement software solutions Source: PUE - A Comprehensive Examination of the Metric
PUE Overview Critical Mechanical Path Measurement Points 19 Efforts should be directed at determining the energy usage by system, including but not limited to the following examples: Cooling plant Chillers Towers Pumps Economizers Thermal storage Secondary chilled-water CRAHs Lighting Fans (fresh air and exhaust) Security Fire suppression systems Source: PUE - A Comprehensive Examination of the Metric
PUE Overview 20 Addressing Challenges thru sub-zones 1: Billing Reconciliation Monitors the building utility metering Point of Entry, providing power, oil, water, and gas consumption information and CO 2 emissions. 2: Switchboard Distribution Board - Monitors sub meters at the main distribution board for DC related equipment. 6: Device Level Monitoring monitoring and control of per outlet or device power, within a data center. 3: Plant Equipment - Distributed monitoring of supporting facility services, including individual chillers, AHU, CRAC, UPS and lighting circuits. 4: Branch Circuit Monitoring of Data Hall Focuses on monitoring of total 1 rack or cabinet IT loads and environmentals. 2 5 6 5: Rack & Cabinet Level Monitoring Intelligent monitoring and control of power 3 4 Power and energy management issues can be identified and resolved throughout all zones
PUE Overview Addressing Challenges thru sub-zones 21 1 Water Power In Oil Gas 2 Main MV / LV Distribution Board (A or B) UPS 3 Lighting Boilers Fire Security Chillers CRAC Back up Generator 4 Power Monitoring Appliances Data Hall Environmental Sub PDU 6 Data Rack or Free Standing Equipment 5 Individual Payloads & Devices
Measure PUE 22 SOURCE ENERGY Energy Type Weighting Factor Electricity 1.00 Natural gas 0.35 Fuel oil 0.35 Other fuels 0.35 District chilled water 0.35 District steam 0.40 Source: PUE - A Comprehensive Examination of the Metric
Measure PUE 23 Source: PUE - A Comprehensive Examination of the Metric
Report PUE 24 Sample PUE Report Interpretation 2.25 PUE L1, Single PUE measurement (2.25) taken using a level 1 meter placement 1.95 PUE L1, YM Yearly average PUE (1.95) using data points gathered monthly with a level 1 meter placement 1.6 PUE L1, MW Monthly average PUE (1.6) using data points gathered weekly with a level 1 meter placement 2.43 PUE L1, WD Weekly average PUE (2.43) using data points gathered daily with a level 1 meter placement 1.8 PUE L2, WC Weekly average PUE using data points gathered continuously with a level 2 meter placement 2.1 PUE L3, YC Yearly average PUE (2.1) using continuous measurements with a level 3 meter placement Source: PUE - A Comprehensive Examination of the Metric
Class of Measurement 25 Class Description Benefit to reporting organization unrecognized A publicly reported result with no claims of following TGG s guidelines. TGG will not comment on unrecognized results Reported Registered Certified A publicly reported result by the reporting organization claiming they followed TGG s measurement recommendations and nomenclature guidelines. TGG will not comment on reported results. A publicly reported result, with key report contextual data provided to TGG by the reporting organization to TGG s data center performance database. A publicly reported results, with key additional data required for third-party validation or certification of results, provided to TGG by the reporting organization. Reporting organization can use standard materials from The Green Grid to explain process and results to audience Official registration of reported results. Receipt of registration number from TGG. Link to public report data from TGG s website. All benefits applicable to registered results, plus, consideration of reported results in future TGG award or recognition program
Agenda: 26 1. Data Center Challenges 2. Power Usage Effectiveness Overview 3. Data Centre Infrastructure Management Software Requirements
DCIM Software Requirements 27 DCIM Software provides: A single reporting platform for multiple data sources across all zones Strong reporting, visualization & analytics Scalable, highly accurate energy, environmental and physical security monitoring Unrivalled levels of granular monitoring with exceptional accuracy Real time dynamic and historical reporting Business intelligence to improve efficiencies and reduce costs Helping you to: Drive operational advantages, sustainability benefits, and optimization programs Reduce OPEX Optimize capacity management Support centralized management and transparency of information Drive efficiency with a single pane of glass
To summarize: PUE is the right metric to measure the efficiency of the data centre DCIM is the way to eliminate risks in Data Center operation by proactive Capacity Planning DCIM supports you in rightsizing your energy equation across Facility and IT Full featured DCIM suites supports your decision making with knowledge rather then information overload
building a smarter, unified business foundation 29