The Ecological Impact of Compute

Slide 1

Slide 1 text

https://upload.wikimedia.org/wikipedia/commons/d/d7/Wikimedia_Foundation_Servers-8055_24.jpg Ecological Compute IMPACT OF THE

Slide 2

Slide 2 text

@SETHVARGO   

Slide 3

Slide 3 text

No content

Slide 4

Slide 4 text

https://upload.wikimedia.org/wikipedia/commons/5/5b/Italian_legal_hardback_books_open.jpg DISCLAIMERS BUT FIRST A FEW

Slide 5

Slide 5 text

https://upload.wikimedia.org/wikipedia/commons/4/4a/March_Against_Monsanto_Vancouver.jpg I AM NOT AN ENVIRONMENTAL ACTIVIST

Slide 6

Slide 6 text

https://upload.wikimedia.org/wikipedia/commons/1/1f/Lincoln_mkc_front.jpg 19 MPG

Slide 7

Slide 7 text

OOPS https://upload.wikimedia.org/wikipedia/commons/8/87/Nest_Learning_Thermostat_(cropped).JPG

Slide 8

Slide 8 text

https://upload.wikimedia.org/wikipedia/commons/5/51/Old_book_-_Basking_Ridge_Historical_Society.jpg BACKGROUND AND NOW SOME

Slide 9

Slide 9 text

https://upload.wikimedia.org/wikipedia/commons/c/c3/Internet_map_4096.png

Slide 10

Slide 10 text

EVERY 60 SECONDS 15,000 MUSIC DOWNLOADS  350,000 TWITTER TWEETS  5 MILLION GOOGLE SEARCHES  204 MILLION EMAIL MESSAGES  1.8 MILLION FACEBOOK LIKES  $272,000 AMAZON PURCHASES 

Slide 11

Slide 11 text

EVERY 60 SECONDS 204 MILLION 5 MILLION 1.8 MILLION 350,000 $272,000 15,000 EMAIL MESSAGES GOOGLE SEARCHES FACEBOOK LIKES TWITTER TWEETS AMAZON PURCHASES MUSIC DOWNLOADS      

Slide 12

Slide 12 text

91 BILLION KILOWATT-HOURS https://static.pexels.com/photos/7000/fre-sonneveld-powerlines.jpg

Slide 13

Slide 13 text

https://upload.wikimedia.org/wikipedia/commons/d/d6/Coal_power_plant_Datteln_2_Crop1.png 500-MEGAWATT POWER PLANTS

Slide 14

Slide 14 text

https://upload.wikimedia.org/wikipedia/commons/d/d6/Coal_power_plant_Datteln_2_Crop1.png 500-MEGAWATT POWER PLANTS 34 TIMES

Slide 15

Slide 15 text

Russia Japan India Germ any Canada France Brazil S. Korea UK Italy Spain Taiwan S. Africa Australia M exico Suppresses China and USA

Slide 16

Slide 16 text

Russia Japan India Germ any Canada France Brazil S. Korea UK Italy Spain Taiwan S. Africa Australia M exico Suppresses China and USA 

Slide 17

Slide 17 text

100 MILLION TONS OF CARBON POLLUTION https://upload.wikimedia.org/wikipedia/commons/c/cb/Automobile_exhaust_gas.jpg

Slide 18

Slide 18 text

100 MILLION TONS OF CARBON POLLUTION https://upload.wikimedia.org/wikipedia/commons/c/cb/Automobile_exhaust_gas.jpg 26.5 MILLION AUTOMOBILES

Slide 19

Slide 19 text

100 MILLION TONS OF CARBON POLLUTION https://upload.wikimedia.org/wikipedia/commons/c/cb/Automobile_exhaust_gas.jpg 26.5 MILLION AUTOMOBILES 470 MILLION TREES

Slide 20

Slide 20 text

https://i.vimeocdn.com/video/526825393_1280x960.jpg SOLUTIONS NOT SO OBVIOUS

Slide 21

Slide 21 text

INCREASE IN PUBLIC CLOUD USE INCREASE IN CARBON FOOTPRINT

Slide 22

Slide 22 text

CORRELATION CAUSATION INCREASE IN PUBLIC CLOUD USE INCREASE IN CARBON FOOTPRINT

Slide 23

Slide 23 text

http://www.thegraphicrecorder.com/wp-content/uploads/2012/01/CorrelationCausationFinal1.jpg

Slide 24

Slide 24 text

4 TYPES OF DATA CENTERS https://upload.wikimedia.org/wikipedia/commons/6/67/Inside_Suite.jpg

Slide 25

Slide 25 text

WHOLESALE CO-LO CUSTOMER

Slide 26

Slide 26 text

WHOLESALE CO-LO RETAIL CO-LO CUSTOMER CUSTOMER 2 CUST 1 C3

Slide 27

Slide 27 text

WHOLESALE CO-LO RETAIL CO-LO MANAGED HOSTING CUSTOMER CUSTOMER 2 CUST 1 C3 CUSTOMER 2 CUST 1 C3

Slide 28

Slide 28 text

WHOLESALE CO-LO RETAIL CO-LO MANAGED HOSTING CLOUD PROVIDER CUSTOMER CUSTOMER 2 CUST 1 C3 CUSTOMER 2 CUST 1 C3 CUSTOMERS

Slide 29

Slide 29 text

WHOLESALE CO-LO RETAIL CO-LO MANAGED HOSTING CLOUD PROVIDER 49% 27% 19% 5%

Slide 30

Slide 30 text

Cloud Provider 5% Managed Hosting 19% Retail Co-lo 27% Wholesale Co-lo 49%

Slide 31

Slide 31 text

CAUSES OF INEFFICIENCY http://www.trbimg.com/img-57322fc1/turbine/ct-tsa-airport-security-lines-20160510

Slide 32

Slide 32 text

OVER-PROVISIONING OF IT RESOURCES https://static.pexels.com/photos/7000/fre-sonneveld-powerlines.jpg #1

Slide 33

Slide 33 text

PEAK PROVISIONING https://upload.wikimedia.org/wikipedia/commons/8/86/Peak_of_the_Matterhorn,_seen_from_Zermatt,_Switzerland.jpg

Slide 34

Slide 34 text

1 2 3 4 5 8 9 10 11 12 7 6 15 16 17 18 19 14 13 22 23 24 25 26 21 20 29 30 28 27 BLACK FRIDAY NEW SERVERS These numbers are ﬁctitious and are not based on statistical evidence; shown for demonstrative purposes only

Slide 35

Slide 35 text

1 2 3 4 5 8 9 10 11 12 7 6 15 16 17 18 19 14 13 22 23 24 25 26 21 20 29 30 28 27 BLACK FRIDAY NEW SERVERS  These numbers are ﬁctitious and are not based on statistical evidence; shown for demonstrative purposes only

Slide 36

Slide 36 text

1 2 3 4 5 8 9 10 11 12 7 6 15 16 17 18 19 14 13 22 23 24 25 26 21 20 29 30 28 27 BLACK FRIDAY NEW SERVERS  – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 These numbers are ﬁctitious and are not based on statistical evidence; shown for demonstrative purposes only

Slide 37

Slide 37 text

1 2 3 4 5 8 9 10 11 12 7 6 15 16 17 18 19 14 13 22 23 24 25 26 21 20 29 30 28 27 BLACK FRIDAY NEW SERVERS  – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 + $10,000 These numbers are ﬁctitious and are not based on statistical evidence; shown for demonstrative purposes only

Slide 38

Slide 38 text

1 2 3 4 5 8 9 10 11 12 7 6 15 16 17 18 19 14 13 22 23 24 25 26 21 20 29 30 28 27 BLACK FRIDAY NEW SERVERS  – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 – $500 + $10,000 – $500 – $500 – $500 – $500 – $500 These numbers are ﬁctitious and are not based on statistical evidence; shown for demonstrative purposes only

Slide 39

Slide 39 text

INVALID IT PROCUREMENT FOCUS #2

Slide 40

Slide 40 text

HOW IT PROCUREMENT DECISIONS SHOULD BE ANALYZED http://www.policonomics.com/wp-content/uploads/Cost-analysis-long-term.jpg

Slide 41

Slide 41 text

HOW IT PROCUREMENT DECISIONS ARE ACTUALLY MADE http://acclinet.com/images/stories/sun-dell-ad.jpg

Slide 42

Slide 42 text

HOW IT PROCUREMENT DECISIONS ARE ACTUALLY MADE https://i.ytimg.com/vi/J2WNUXTRNes/maxresdefault.jpg

Slide 43

Slide 43 text

HOW IT PROCUREMENT DECISIONS ARE ACTUALLY MADE http://ad-design.966v.com/static_images/20160724/cd767d958d873b6ab96082b93951e53865607b3ea4ca315e0aeff539.jpg

Slide 44

Slide 44 text

UNUSED "ZOMBIE" SERVERS https://upload.wikimedia.org/wikipedia/commons/1/17/2007-04-07_-_London_-_Flashmob_-_Fleshmob_-_Zombie_Walk_-_Zombies_(4889850244).jpg #3

Slide 45

Slide 45 text

An estimated 20-30% of servers in [wholesale] data centers today are idle, obsolete, or unused but are still plugged in and consuming energy doing nothing.

Slide 46

Slide 46 text

https://www.youtube.com/watch?v=95y0tPseZRE

Slide 47

Slide 47 text

IT managers cannot identify owners for 15-30% of the servers, but are reluctant to decommission equipment, fearing potential impacts on business or application functions.

Slide 48

Slide 48 text

https://journal.uptimeinstitute.com/wp-content/uploads/2014/08/server-roundup-cover-image.jpg

Slide 49

Slide 49 text

https://journal.uptimeinstitute.com/wp-content/uploads/2014/08/server-roundup-cover-image.jpg DECOMMISSIONED MORE THAN 9,500 SERVERS $5M IN ANNUAL ENERGY SAVINGS 5 MEGAWATTS IN IT LOAD 4 MEGAWATTS OF ASSOCIATED COOLING

Slide 50

Slide 50 text

https://journal.uptimeinstitute.com/wp-content/uploads/2014/08/server-roundup-cover-image.jpg MISALIGNMENT OF RESPONSIBILITIES ELECTRIC COSTS PAID BY A DIFFERENT DEPARTMENT NO INCENTIVE FOR DATA CENTER TEAMS TO REDUCE LOAD

Slide 51

Slide 51 text

https://upload.wikimedia.org/wikipedia/commons/d/df/Port_Authority_bus_Pittsburgh_3216.jpg LACK OF STANDARD UTILIZATION METRIC #4

Slide 52

Slide 52 text

80% CPU 

Slide 53

Slide 53 text

80% CPU 60% MEM  

Slide 54

Slide 54 text

80% CPU 60% MEM 1500 IOPS   

Slide 55

Slide 55 text

80% CPU 60% MEM 1500 IOPS ?    × ÷ ≈

Slide 56

Slide 56 text

http://i.huffpost.com/gen/1821469/images/o-LANGUAGES-facebook.jpg

Slide 57

Slide 57 text

https://upload.wikimedia.org/wikipedia/commons/4/43/Okinawan_High_Schools_compete_in_race_at_MCAS_Futenma_141001-M-PU373-021.jpg COMPETING PRIORITIES FOR EFFICIENCY #5

Slide 58

Slide 58 text

Efﬁciency 4% Deployments 24% Security 24% Reliability 24% Availability 24%

Slide 59

Slide 59 text

Deployments 25% Security 25% Reliability 25% Availability 25%

Slide 60

Slide 60 text

http://gssolutions.com/wp-content/uploads/2014/12/cloudit.jpg CLOUD TO THE

Slide 61

Slide 61 text

No content

Slide 62

Slide 62 text

WE DON'T GET THE SAME PERFORMANCE ON THE CLOUD MANY ORGANIZATIONS DO NOT UNDERSTAND THEIR APPLICATION REQUIREMENTS MYTH FACT

Slide 63

Slide 63 text

THE CLOUD IS INSECURE – WE HAVE TO BE HIPPA/PCI COMPLIANT CLOUD PROVIDERS PAY MILLIONS OF DOLLARS AND HIRE THE MOST EXPERIENCED PROFESSIONALS ON THESE TOPICS. ARE YOU BETTER IF COMPLIANCE IS SOMEONE'S PART-TIME JOB? MYTH FACT

Slide 64

Slide 64 text

THE CLOUD IS TOO EXPENSIVE FOR MOST USE CASES, IT'S PROBABLY CHEAPER AND MORE FLEXIBLE MYTH FACT

Slide 65

Slide 65 text

WE RUN ON OUR OWN BARE METAL AND ACHIEVE 80% UTILIZATION USING A SCHEDULER AND ROUTINELY AUDIT SERVER USAGE USING AN INVENTORY MANAGEMENT SYSTEM. THIS TALK IS NOT FOR YOU, BUT MYTH FACT

Slide 66

Slide 66 text

I AM ENERGY AND RESOURCE EFFICIENT BECAUSE I AM USING A CLOUD PROVIDER YOU ARE IMPOSING LESS OF AN ENVIRONMENTAL IMPACT, BUT YOU MAY NOT BE USING YOUR POOL OF RESOURCES AT BEST CAPACITY MYTH FACT

Slide 67

Slide 67 text

http://i.huffpost.com/gen/1821469/images/o-LANGUAGES-facebook.jpg LIVE DEMO

Slide 68

Slide 68 text

https://upload.wikimedia.org/wikipedia/commons/8/82/Cloture_grande_faune_A73A.jpg UTILIZATION MAXIMIZING RESOURCE

Slide 69

Slide 69 text

HUMANS MAXIMIZE TIME WITH THESE https://upload.wikimedia.org/wikipedia/commons/0/01/February_calendar.jpg

Slide 70

Slide 70 text

sched·ul·er (human) https://upload.wikimedia.org/wikipedia/commons/0/01/February_calendar.jpg a person or machine that organizes or maintains schedules

Slide 71

Slide 71 text

sched·ul·er (computer) https://upload.wikimedia.org/wikipedia/commons/0/01/February_calendar.jpg a program that arranges computer operations into an appropriate sequence

Slide 72

Slide 72 text

https://cdn.shopify.com/s/ﬁles/1/0167/3936/ﬁles/04-MS-Excel-4-0-Office-4-0_large.gif

Slide 73

Slide 73 text

  Operator Datacenter

Slide 74

Slide 74 text

  Operator Datacenter Skywalker Vader Leia Solo

Slide 75

Slide 75 text

  Operator Datacenter PYTHON PYTHON GOLANG GOLANG GOLANG Skywalker Vader Leia Solo

Slide 76

Slide 76 text

  Operator Datacenter RUBY PYTHON PYTHON PYTHON GOLANG GOLANG GOLANG GOLANG NODE Skywalker Vader Leia Solo

Slide 77

Slide 77 text

  Operator Datacenter RUBY PYTHON PYTHON PYTHON GOLANG GOLANG GOLANG GOLANG NODE Skywalker Vader Leia Solo RUBY VADER LEIA SOLO 192.168.1.4 192.168.1.5 192.168.1.7 192.168.1.253 88:45:13:B6:87:C4 94:CE:4F:C8:54:C3 CA:9A:3D:7F:8B:CB 72:30:9C:0D:1E:74 Randomly kills applications

Slide 78

Slide 78 text

  Operator Datacenter RUBY PYTHON PYTHON PYTHON GOLANG GOLANG GOLANG GOLANG NODE Skywalker Leia Solo RUBY VADER LEIA SOLO 192.168.1.4 192.168.1.5 192.168.1.7 192.168.1.253 88:45:13:B6:87:C4 94:CE:4F:C8:54:C3 CA:9A:3D:7F:8B:CB 72:30:9C:0D:1E:74 Randomly kills applications F F Vader

Slide 79

Slide 79 text

Slide 80

Slide 80 text

  Operator Datacenter RUBY GOLANG GOLANG GOLANG GOLANG NODE Skywalker Leia Solo RUBY VADER LEIA SOLO 192.168.1.4 192.168.1.5 192.168.1.7 192.168.1.253 88:45:13:B6:87:C4 94:CE:4F:C8:54:C3 CA:9A:3D:7F:8B:CB 72:30:9C:0D:1E:74 Randomly kills applications Vader PYTHON PYTHON PYTHON

Slide 81

Slide 81 text

  Operator Datacenter RUBY GOLANG GOLANG GOLANG GOLANG NODE Skywalker Leia Solo RUBY VADER LEIA SOLO 192.168.1.4 192.168.1.9 192.168.1.7 192.168.1.253 88:45:13:B6:87:C4 94:CE:4F:C8:54:C3 CA:9A:3D:7F:8B:CB 72:30:9C:0D:1E:74 Rebuilt on 04/20/2016 Vader PYTHON PYTHON PYTHON

Slide 82

Slide 82 text

Slide 83

Slide 83 text

THIS DOES NOT SCALE http://www.public-domain-image.com/free-images/objects/spring-scale-used-to-determine-ones-weight-in-pounds.jpg

Slide 84

Slide 84 text

Type Work Resources CPU Scheduler Threads Physical Cores EC2 / Nova Virtual Machines Hypervisors Hadoop YARN MapReduce Jobs Client Nodes Cluster Scheduler Applications Machines https://upload.wikimedia.org/wikipedia/commons/a/a0/Cedar_Mtn_Wilderness.jpg

Slide 85

Slide 85 text

No content

Slide 86

Slide 86 text

http://i.huffpost.com/gen/1821469/images/o-LANGUAGES-facebook.jpg ANOTHER LIVE DEMO

Slide 87

Slide 87 text

Higher Resource Utilization Decouple Work from Resources Better Quality of Service ADVANTAGES https://upload.wikimedia.org/wikipedia/commons/3/38/GiantEagleStowOhio.jpg

Slide 88

Slide 88 text

Higher Resource Utilization ADVANTAGES https://upload.wikimedia.org/wikipedia/commons/3/38/GiantEagleStowOhio.jpg Bin Packing Over-Subscription Job Queueing

Slide 89

Slide 89 text

Decouple Work from Resources ADVANTAGES https://upload.wikimedia.org/wikipedia/commons/3/38/GiantEagleStowOhio.jpg Abstraction API Contracts Standardization

Slide 90

Slide 90 text

Better Quality of Service ADVANTAGES https://upload.wikimedia.org/wikipedia/commons/3/38/GiantEagleStowOhio.jpg Priorities Resource Isolation Pre-emption

Slide 91

Slide 91 text

https://upload.wikimedia.org/wikipedia/commons/8/85/Trail_Ridge_Road_and_Longs_Peak_by_RO.jpg SCALABILITY BONUS

Slide 92

Slide 92 text

Nomad Million Container Challenge 1,000 Jobs 1,000 Tasks per Job 5,000 Hosts on GCE 1,000,000 Containers

Slide 93

Slide 93 text

No content

Slide 94

Slide 94 text

http://i.huffpost.com/gen/1821469/images/o-LANGUAGES-facebook.jpg No one will ever need to schedule a million containers.

Slide 95

Slide 95 text

https://upload.wikimedia.org/wikipedia/commons/thumb/e/e1/Bill_Gates,_WEF_2009_Davos.jpg/1280px-Bill_Gates,_WEF_2009_Davos.jpg 640K ought to be enough for anybody

Slide 96

Slide 96 text

No content

Slide 97

Slide 97 text

No content

Slide 98

Slide 98 text

No content

Slide 99

Slide 99 text

https://upload.wikimedia.org/wikipedia/commons/d/d7/Wikimedia_Foundation_Servers-8055_24.jpg 1. USE A CLOUD PROVIDER 2. USE A SCHEDULER 3. MEASURE EVERYTHING

Slide 100

Slide 100 text

@SETHVARGO    QUESTIONS?