SlideShare a Scribd company logo
Data Center Efficiency
       The Google way.



Arjun Mohan
Hima Sudarsan
Nithin Madhavan
Surej Shams
Background
 Major data centers require a large
  amount of energy to power and maintain
  them.
 The cost of running large data centers is a
  significant component of the overall IT
  budget of firms.
 There are two components to the energy
  cost of data centers:
   Cost of running computers.
   Cost of cooling them.
 In 2011, the EPA estimated that data centers
  in the U.S. used about 12 GW of
  power, equivalent to the output of 25 major
  power plants.
 Due to increasing power demands, by
  2020, the world's computer servers will match
  or exceed the carbon emissions of the airline
  industry.
Ways to go Green
  Methods to reduce power consumption.
 Distribute data centers worldwide, placing
  them in areas where it‟s more efficient to
  power them.
 Use of hydropower.
 Virtualization.
 Google is an industry leader in data
  center efficiency.
 It builds its own custom data centers
  rather than rely on standard industry
  equipment and practices.
Q. 1

What is PUE, and why is it an important
place to start when considering how to
reduce data center power consumption?
What value of PUE should data center
managers strive for?
• Power Usage Effectiveness (PUE) is a
measure of               how efficiently a
computer data center uses its power.
• It is the ratio of total amount of power
used by a computer data center facility to
the power delivered to computing
equipment.
• PUE was developed by a consortium
called The Green Grid.
• The
    data center industry uses PUE to
measure efficiency.

•It help reduce the energy used for non-
computing functions like cooling and power
distribution.

•A PUE of 2.0 means that for every watt of IT
power, an additional watt is consumed to cool
and distribute power to the IT equipment.
 A PUE of 1.0 is considered to be ideal.
 A PUE closer to 1.0 means nearly all of the
  energy is used for computing.
 For using PUE effectively, it's important to
  measure often.
 It should be sampled at least once per
  second.
Q. 2
Describe the five methods recommended by
Google for reducing power consumption.
Google recommends five methods for
reducing power consumption:
1. Measure PUE
2. Manage airflow
3. Adjust thermostat
4. Use free cooling
5. Optimize power distribution
1.Measure PUE
•   Power usage effectiveness (PUE) is
   a measure of how efficiently a
   computer data center uses its power.
• PUE measures the amount of non-
   computing overhead energy to the
   amount of energy used to power the
   actual machines.
2. Manage airflow
• Good air flow is fundamental to efficient
   data center operation.
• For healthy server and computer function
   the internal temperature of the data
   center is to be kept at the right level.
• First thermal modeling is used to locate
   hot spots and better understand the air
   flow in data centers.
• Then simple, cost-effective tactics to
   prevent hot air from mixing with the cold.
 Use blanking panels (or flat sheets of
  metal) to close off empty rack slots and
  prevent hot aisle air from seeping into to
  the cold aisle.
 Hang plastic curtains (like those used in
  commercial refrigerators) to seal off the
  cold aisle.
 Enclose areas with components that run
  hotter (such as power supply units or
  PSUs) with plastic curtains.
3. Adjust thermostat
• A thermostat is a control unit for a heating
or cooling system or a component part of a
heater or air conditioner
• Senses the temperature of a system so that
the system's temperature is maintained near a
desired set point.
• The need for energy intensive air
conditioning is eliminated by raising the cold
aisle of the data center to 80°F or higher.
• This helps to reduce energy use significantly.
4.Use free cooling
•Free cooling is removing heat from a data
center without using a large air-
conditioner, or chiller.
• Cold outside air, water evaporation, or
thermal reservoirs are used to cool the
equipment.
• Avoiding the need for mechanical chillers is
one of the best way for energy and cost
savings.
5. Optimize power distribution
•   To optimize power distribution, Google
    use high-efficiency power supplies,
    minimize the number of times power is
    being converted from one type of
    electrical current to another, and keep
    power supplies as close to the load as
    possible.
•   The servers lose only a little over 15% of
    the electricity they pull from the wall
    during power conversion steps—less than
    half of what is lost in a typical server.
Q. 3

How much of the world‟s global
greenhouse gases are the result of
computing? List several factors that
will tend to retard or accelerate data
center power consumption both globally
and in the United States.
   Around 2% of the global greenhouse gas
    emissions are a result of computing.
   In this 2%, 15% constitutes data centers
    and 5 % includes large internet data
    centers.
Factors that accelerate power
consumption

   Advancements in IT.
   Cloud computing.
   Audio and Video streaming.
Factors that reduce power consumption
   Air management

   Chillers
   Virtualization.

     This technique allows servers to perform
    multiple tasks at once.

   Distributing datacenters worldwide.
Q. 4
 Data center power consumption will be
  high in developed countries, because most
  of the major data centers are located in
  these countries.
 Now, to reduce data center power
  consumption, companies are preferring
  places with lower ambient temperatures.
  This includes, Iceland and Scandinavia.
   This map shows Google‟s data center
    locations worldwide.
 The map shows that a large number of
  data centers are located in North
  America and Europe.
 This will include most of the developed
  countries.
Q. 5
   Imagine that a company has developed an
    advanced technology that allows it to
    reduce its data center requirements by an
    unprecedented amount, and creates a
    competitive advantage for the company in
    the data center market. Why should it
    share that technology with other data
    center firms?
 If a firm share its techniques, the rest of
  the industry can also be more efficient
  centers, and decrease global emissions of
  green house gases.
 Google is considered to be an industry
  leader in data center efficiency.
 It builds its own custom data centers.
 But Google will not share its
  virtualization technology.
   Google takes many other steps to share
    its energy efficient best practices with
    other firms and the public.
Q. 6
   Should the government or an industry
    association regulate the carbon emissions
    of the data center industry as they do the
    airline industry? Or the automobile
    industry? Or is it sufficient to rely on the
    competitive market place to drive down
    energy consumption in data centers?
 A government or an industry association
  should regulate the carbon emissions of
  the data center industry as they do for
  other industries.
 Such an association can make the
  companies aware about the importance of
  „going green‟ and there by reducing
  emissions.
 The Environmental Protection Agency, In
  2007, projected that the power
  requirements of data centers would double
  from 2005 to 2010.
 There is plenty of incentive for online e-
  commerce companies to reduce their data
  center power consumption.
 In addition, competition among the industry‟s
  largest providers of data centers are
  incentivized to help their customers “go
  green” and reduce costs.
THANK YOU

More Related Content

PPTX
Data centers
PPTX
Facebook Data Center
PPTX
Data Center Efficiency Workshop 2016
PPTX
Proactively Managing Your Data Center Infrastructure
PPTX
Essential elements of data center operations
PPTX
Data Center Cooling Design - Datacenter-serverroom
PDF
Schneider Electric Data Center Reference Designs
PDF
Management Information System for BCA
Data centers
Facebook Data Center
Data Center Efficiency Workshop 2016
Proactively Managing Your Data Center Infrastructure
Essential elements of data center operations
Data Center Cooling Design - Datacenter-serverroom
Schneider Electric Data Center Reference Designs
Management Information System for BCA

What's hot (20)

PPTX
4 steps to quickly improve pue through airflow management
PPT
Simplifying Data Center Design/ Build
PPTX
Green data centers
PPT
Chapter 6
PPTX
Data Center Tiers Explained
PDF
Principles of Risk Management
PDF
Datacenter Strategy, Design, and Build
PDF
Multi tiered hybrid data center design
PPTX
Data center maintenance
PDF
Top 10 Data Center Success Criteria
PPTX
Gaining Data Center Cooling Efficiency Through Airflow Management
PPTX
HVAC for Data Centers
DOC
Data Center Proposal (System Network Administration)
PPTX
Modular Data Center Design
PPT
Management Information System [Kenneth Laudon]
PDF
ASHRAE TC9.9 - DATA CENTER STANDARD AND BEST PRACTICES
ODP
Server room presentation 16th january 2014
PPTX
Data Centre Design Guideline and Standards
PPTX
Dell case study final presentation
PPT
Laudon mis14 ch02
4 steps to quickly improve pue through airflow management
Simplifying Data Center Design/ Build
Green data centers
Chapter 6
Data Center Tiers Explained
Principles of Risk Management
Datacenter Strategy, Design, and Build
Multi tiered hybrid data center design
Data center maintenance
Top 10 Data Center Success Criteria
Gaining Data Center Cooling Efficiency Through Airflow Management
HVAC for Data Centers
Data Center Proposal (System Network Administration)
Modular Data Center Design
Management Information System [Kenneth Laudon]
ASHRAE TC9.9 - DATA CENTER STANDARD AND BEST PRACTICES
Server room presentation 16th january 2014
Data Centre Design Guideline and Standards
Dell case study final presentation
Laudon mis14 ch02
Ad

Viewers also liked (20)

PPTX
Google data centers
PDF
Google's internal systems
PPTX
POWER POINT PRESENTATION ON DATA CENTER
PPTX
Communication networks at google
PPTX
Datacenter overview
PDF
All about Google
PPTX
Big data ppt
PDF
How Digital drives auto shopping | Google Think Insight
PPTX
Google search engine system analysis
ODP
The Age of Collaboration
PPTX
Waste Not Want Not
PPTX
IBM Counter Fraud Management
PDF
A Look Inside Google’s Data Center Networks
PPT
Big Brands & Facebook: Demographics, Case Studies & Best Practices
PDF
OCP Switch Overview
DOCX
SRUTHY RK.KSE IRINJALAKUDA ORGANIZATION STUDY
PPT
Competitive strategies in different types of industries
DOC
Os mariya jasmine 2012
PPTX
MIS presentation on Skype Business Model
Google data centers
Google's internal systems
POWER POINT PRESENTATION ON DATA CENTER
Communication networks at google
Datacenter overview
All about Google
Big data ppt
How Digital drives auto shopping | Google Think Insight
Google search engine system analysis
The Age of Collaboration
Waste Not Want Not
IBM Counter Fraud Management
A Look Inside Google’s Data Center Networks
Big Brands & Facebook: Demographics, Case Studies & Best Practices
OCP Switch Overview
SRUTHY RK.KSE IRINJALAKUDA ORGANIZATION STUDY
Competitive strategies in different types of industries
Os mariya jasmine 2012
MIS presentation on Skype Business Model
Ad

Similar to Google ppt. mis (20)

PPTX
Green computing
PPT
Green data center_rahul ppt
PDF
Express Computer
PPTX
Green Data Centre for banks
PPTX
Data centerefficiency
PDF
Improvements in Data Center Management
PPTX
Green computing 1 1
PPT
The next wave of GreenIT
PDF
Improve sustainability through energy insights - Infographic
PDF
Motivation for Green Computing, an Analytical Approach
PDF
Full chapter in a single perfect format 2
PPTX
It&smart grid
PPT
PPT
Optimizing The Data Centre Environment
PPSX
Green computing
PPTX
The next hope of future is a green computing
DOCX
Datacenter ISO50001 and CoC
PDF
Green it at university of bahrain
Green computing
Green data center_rahul ppt
Express Computer
Green Data Centre for banks
Data centerefficiency
Improvements in Data Center Management
Green computing 1 1
The next wave of GreenIT
Improve sustainability through energy insights - Infographic
Motivation for Green Computing, an Analytical Approach
Full chapter in a single perfect format 2
It&smart grid
Optimizing The Data Centre Environment
Green computing
The next hope of future is a green computing
Datacenter ISO50001 and CoC
Green it at university of bahrain

Google ppt. mis

  • 1. Data Center Efficiency The Google way. Arjun Mohan Hima Sudarsan Nithin Madhavan Surej Shams
  • 2. Background  Major data centers require a large amount of energy to power and maintain them.  The cost of running large data centers is a significant component of the overall IT budget of firms.  There are two components to the energy cost of data centers: Cost of running computers. Cost of cooling them.
  • 3.  In 2011, the EPA estimated that data centers in the U.S. used about 12 GW of power, equivalent to the output of 25 major power plants.  Due to increasing power demands, by 2020, the world's computer servers will match or exceed the carbon emissions of the airline industry.
  • 4. Ways to go Green Methods to reduce power consumption.  Distribute data centers worldwide, placing them in areas where it‟s more efficient to power them.  Use of hydropower.  Virtualization.
  • 5.  Google is an industry leader in data center efficiency.  It builds its own custom data centers rather than rely on standard industry equipment and practices.
  • 6. Q. 1 What is PUE, and why is it an important place to start when considering how to reduce data center power consumption? What value of PUE should data center managers strive for?
  • 7. • Power Usage Effectiveness (PUE) is a measure of how efficiently a computer data center uses its power. • It is the ratio of total amount of power used by a computer data center facility to the power delivered to computing equipment. • PUE was developed by a consortium called The Green Grid.
  • 8. • The data center industry uses PUE to measure efficiency. •It help reduce the energy used for non- computing functions like cooling and power distribution. •A PUE of 2.0 means that for every watt of IT power, an additional watt is consumed to cool and distribute power to the IT equipment.
  • 9.  A PUE of 1.0 is considered to be ideal.  A PUE closer to 1.0 means nearly all of the energy is used for computing.  For using PUE effectively, it's important to measure often.  It should be sampled at least once per second.
  • 10. Q. 2 Describe the five methods recommended by Google for reducing power consumption.
  • 11. Google recommends five methods for reducing power consumption: 1. Measure PUE 2. Manage airflow 3. Adjust thermostat 4. Use free cooling 5. Optimize power distribution
  • 12. 1.Measure PUE • Power usage effectiveness (PUE) is a measure of how efficiently a computer data center uses its power. • PUE measures the amount of non- computing overhead energy to the amount of energy used to power the actual machines.
  • 13. 2. Manage airflow • Good air flow is fundamental to efficient data center operation. • For healthy server and computer function the internal temperature of the data center is to be kept at the right level. • First thermal modeling is used to locate hot spots and better understand the air flow in data centers. • Then simple, cost-effective tactics to prevent hot air from mixing with the cold.
  • 14.  Use blanking panels (or flat sheets of metal) to close off empty rack slots and prevent hot aisle air from seeping into to the cold aisle.  Hang plastic curtains (like those used in commercial refrigerators) to seal off the cold aisle.  Enclose areas with components that run hotter (such as power supply units or PSUs) with plastic curtains.
  • 15. 3. Adjust thermostat • A thermostat is a control unit for a heating or cooling system or a component part of a heater or air conditioner • Senses the temperature of a system so that the system's temperature is maintained near a desired set point. • The need for energy intensive air conditioning is eliminated by raising the cold aisle of the data center to 80°F or higher. • This helps to reduce energy use significantly.
  • 16. 4.Use free cooling •Free cooling is removing heat from a data center without using a large air- conditioner, or chiller. • Cold outside air, water evaporation, or thermal reservoirs are used to cool the equipment. • Avoiding the need for mechanical chillers is one of the best way for energy and cost savings.
  • 17. 5. Optimize power distribution • To optimize power distribution, Google use high-efficiency power supplies, minimize the number of times power is being converted from one type of electrical current to another, and keep power supplies as close to the load as possible. • The servers lose only a little over 15% of the electricity they pull from the wall during power conversion steps—less than half of what is lost in a typical server.
  • 18. Q. 3 How much of the world‟s global greenhouse gases are the result of computing? List several factors that will tend to retard or accelerate data center power consumption both globally and in the United States.
  • 19. Around 2% of the global greenhouse gas emissions are a result of computing.  In this 2%, 15% constitutes data centers and 5 % includes large internet data centers.
  • 20. Factors that accelerate power consumption  Advancements in IT.  Cloud computing.  Audio and Video streaming.
  • 21. Factors that reduce power consumption  Air management  Chillers  Virtualization. This technique allows servers to perform multiple tasks at once.  Distributing datacenters worldwide.
  • 22. Q. 4
  • 23.  Data center power consumption will be high in developed countries, because most of the major data centers are located in these countries.  Now, to reduce data center power consumption, companies are preferring places with lower ambient temperatures. This includes, Iceland and Scandinavia.
  • 24. This map shows Google‟s data center locations worldwide.
  • 25.  The map shows that a large number of data centers are located in North America and Europe.  This will include most of the developed countries.
  • 26. Q. 5  Imagine that a company has developed an advanced technology that allows it to reduce its data center requirements by an unprecedented amount, and creates a competitive advantage for the company in the data center market. Why should it share that technology with other data center firms?
  • 27.  If a firm share its techniques, the rest of the industry can also be more efficient centers, and decrease global emissions of green house gases.  Google is considered to be an industry leader in data center efficiency.  It builds its own custom data centers.  But Google will not share its virtualization technology.
  • 28. Google takes many other steps to share its energy efficient best practices with other firms and the public.
  • 29. Q. 6  Should the government or an industry association regulate the carbon emissions of the data center industry as they do the airline industry? Or the automobile industry? Or is it sufficient to rely on the competitive market place to drive down energy consumption in data centers?
  • 30.  A government or an industry association should regulate the carbon emissions of the data center industry as they do for other industries.  Such an association can make the companies aware about the importance of „going green‟ and there by reducing emissions.
  • 31.  The Environmental Protection Agency, In 2007, projected that the power requirements of data centers would double from 2005 to 2010.  There is plenty of incentive for online e- commerce companies to reduce their data center power consumption.  In addition, competition among the industry‟s largest providers of data centers are incentivized to help their customers “go green” and reduce costs.