Keynote: Building IT Green: A Journey of Platforms, Data, and Developer Empowerment at Deutsche Bahn

CNCF [Cloud Native Computing Foundation]
21 Mar 202417:14

Summary

TLDRThe speaker shares a personal story of struggling with illness while highlighting the urgent issue of climate change. They emphasize the importance of individual and collective action, drawing parallels with the power of a surfing wave. The narrative then shifts to the role of technology and digitalization in addressing climate challenges, particularly focusing on sustainable mobility and the green digitalization initiative. The speaker advocates for the empowerment of developers to make significant changes through efficient use of resources and the implementation of energy-saving strategies in cloud operations, exemplified by the use of Kubernetes and other tools. The message concludes with a call to action for immediate and informed engagement in environmental conservation efforts.

Takeaways

  • ๐ŸŒŠ The speaker uses the metaphor of creating a 'big wave' to illustrate the collective power and energy needed to address climate change.
  • ๐Ÿ˜ท The speaker shares a personal struggle with flu, using it as a narrative to emphasize the importance of overcoming challenges, like tackling climate change.
  • ๐Ÿ–๏ธ A connection to nature through surfing is highlighted as a pivotal moment for the speaker's environmental awareness, emphasizing the need for sustainable practices.
  • ๐ŸŒฑ The urgency of climate change is discussed, noting that environmental issues have been recognized since the '90s, but insufficient political action has been taken.
  • ๐Ÿš„ Sustainable mobility, particularly through train travel, is presented as a crucial solution for achieving climate targets, over relying on infrastructure expansion.
  • ๐Ÿ’ก The impact of digitalization on energy consumption is acknowledged, with a prediction that it will require the electrical consumption equivalent of Sweden or Germany.
  • ๐ŸŒฟ The concept of 'green digitalization' is introduced, aiming to address the energy demands of digital technologies through initiatives supported by CEOs and grassroots movements.
  • ๐Ÿ”ง Developers are identified as key agents of change, with the power to influence software and infrastructure efficiency through their everyday decisions.
  • ๐Ÿค– Kubernetes is praised as a tool for efficient cloud operations and carbon reduction, with its potential to increase resource utilization and reduce waste.
  • ๐Ÿ“ˆ The importance of data visibility and real-time information is emphasized for understanding the energy impact of code changes and workloads.
  • ๐Ÿš€ The call to action is for immediate, small-scale beginnings to harness the intrinsic motivation of developers and communities to care for the planet and reduce carbon footprints.

Q & A

  • What is the main issue the speaker is struggling with at the beginning of the transcript?

    -The speaker is struggling with the flu and its effects, which include fever and hallucination dreams, making them question their ability to stand on stage and give a 15-minute talk.

  • What does the speaker compare to the experience of surfing?

    -The speaker compares the adrenaline and energy rush of catching a wave while surfing to the energy and focus they want from the audience during their talk.

  • What is the urgency the speaker refers to in relation to climate change?

    -The urgency the speaker refers to is the need for immediate action on climate change, as environmental issues were already being discussed in the '90s, and despite some political efforts like the Kyoto Protocol, they were insufficient to address the issue.

  • What is the speaker's connection to the company they mention?

    -The speaker is connected to the company through their role in promoting sustainable mobility, specifically through train travel, which they believe is an important aspect of achieving climate targets.

  • How does the speaker describe the impact of digitalization on energy consumption?

    -The speaker describes the impact of digitalization as having an exponential growth in energy consumption, with estimates suggesting a need for the electricity consumption equivalent to that of Sweden or Germany within the next three years.

  • What is the green digitalization initiative mentioned by the speaker?

    -The green digitalization initiative is a movement started in 2022 aimed at addressing the energy demands and environmental impact of digitalization. It is supported by CEOs and CIOs and has resulted in grassroots efforts from developers within the company.

  • What is the speaker's view on the concept of carbon compensation?

    -The speaker believes that while carbon compensation is an important measure, it cannot be the sole solution due to the exponential growth of energy demands. They argue that there are physical limits to growth and that strategies for sufficiency and efficiency are needed.

  • How does the speaker describe the role of developers in addressing climate change?

    -The speaker views developers as everyday decision-makers when it comes to software and believes that engaging them and providing the right tools is crucial for making changes in how code is developed and infrastructure is managed.

  • What is the significance of Kubernetes for the speaker and their company?

    -Kubernetes is seen as a tool for efficient cloud operations and carbon reduction. The speaker believes it is not just a container orchestration tool but also a platform-building tool that can help achieve high density of containers and improve resource utilization.

  • What solution does the speaker propose for better resource utilization in cloud environments?

    -The speaker proposes the use of vertical pod autoscaling to optimize container workloads according to their needs, which can help reduce wasted resources and improve overall efficiency.

  • How does the speaker suggest measuring the energy impact of workloads?

    -The speaker suggests using tools like Kepler to gather energy metrics and provide real-time data on the energy usage of individual components. This data can then be visualized in dashboards, allowing for better understanding and management of energy consumption.

  • What is the speaker's final advice to the audience?

    -The speaker advises the audience to start taking action now, as the issue of climate change is urgent. They encourage starting small and focusing on empowering developers and leveraging platforms to make a positive impact on the environment.

Outlines

00:00

๐ŸŒŠ Surfing and the Power of Unity

The speaker begins by sharing a personal story of struggling with the flu and feeling overwhelmed at the prospect of speaking to a large audience. However, they find inspiration in their love for surfing, using the metaphor of creating a 'big wave' to illustrate the collective power of the audience. The speaker emphasizes the importance of individual and collective action, particularly in the face of urgent environmental issues like climate change. They recount the early days of environmental awareness in the '90s and the insufficient political actions taken, such as the Kyoto Protocol, highlighting the need for more robust and engaged efforts to address climate change.

05:00

๐ŸŒฑ Green Digitalization and Developer Empowerment

The speaker transitions into discussing the role of digitalization in sustainability, particularly in the context of cloud computing and its increasing energy demands. They express concern over the limitations of current tools for measuring carbon emissions accurately, especially at the application level. The speaker then introduces the concept of 'green digitalization,' initiated by Deustche Telekom in 2022, which aims to tackle the environmental impact of digital technologies. They stress the importance of empowering developers to make informed decisions about resource usage and efficiency, advocating for platforms like Kubernetes as a means to achieve greener cloud operations.

10:02

๐Ÿš‚ Optimizing Cloud Operations for Sustainability

The speaker delves deeper into the practical aspects of optimizing cloud operations for sustainability. They discuss the challenges of managing resource allocation in containerized environments, highlighting the issue of underutilized CPU resources. The speaker advocates for the use of Kubernetes features like vertical pod autoscaling to optimize resource usage and reduce waste. They also mention the benefits of scheduling workloads to align with human work patterns, such as using down scalers to shut down unnecessary processes during off-hours. The speaker emphasizes the need for better data on energy consumption to inform decision-making and introduces tools like Kepler for energy metric analysis, which can provide insights down to the container level.

15:03

๐ŸŒ Community Building and the Future of Green Technology

In the final paragraph, the speaker reflects on the broader implications of the community's role in driving sustainable practices within technology. They stress the urgency of starting now and the importance of data in guiding efforts to reduce energy consumption. The speaker shares their experience with rolling out Kepler, highlighting the challenges of enterprise readiness and the need to balance detailed metrics with practical monitoring. They conclude by encouraging the audience to start small and to leverage the intrinsic motivation of developers to foster a culture of sustainability. The speaker's call to action is to focus on empowerment, community engagement, and the continuous improvement of green technologies.

Mindmap

Keywords

๐Ÿ’กFlu

In the context of the video, 'flu' refers to a contagious viral illness that affects the respiratory system. The speaker mentions having caught the flu over the weekend, which highlights their personal struggle and determination to still deliver the talk despite feeling unwell. This term is used to establish a relatable human element at the beginning of the transcript, setting the stage for the audience to empathize with the speaker's situation.

๐Ÿ’กSurfing

Surfing is a water sport that involves riding on the forward or face of a moving wave, which the speaker uses as a metaphor for the energy and action needed to address climate change. The speaker's first contact with nature and the adrenaline-filled moments experienced while surfing are contrasted with the current urgent need for collective action on climate issues. This keyword ties the personal experience of the speaker to the broader environmental message of the talk.

๐Ÿ’กClimate Change

Climate change is the long-term alteration of temperature and typical weather patterns in a place. In the video, it represents the central theme and urgency that the speaker is trying to convey. The speaker discusses the history of climate change awareness and the need for immediate, collective action to mitigate its effects. This keyword encapsulates the global environmental challenge that the speaker is addressing and calls for action from every individual and organization.

๐Ÿ’กSustainability

Sustainability refers to the practice of meeting current needs without compromising the ability of future generations to meet their own needs. In the context of the video, the speaker emphasizes sustainable mobility, particularly through train travel, as a means to reduce carbon emissions and achieve climate targets. This keyword is integral to the speaker's argument for a shift in our approach to transportation and infrastructure development.

๐Ÿ’กDigitalization

Digitalization is the process of changing from analog or manual systems to digital or computerized systems. The speaker discusses the impact of digitalization on energy consumption and the need for 'green digitalization' initiatives to address the exponential growth in energy demands. This keyword is crucial in understanding the dual role of technology in both contributing to and mitigating environmental issues.

๐Ÿ’กKubernetes

Kubernetes is an open-source platform used for automating deployment, scaling, and management of containerized applications. In the video, the speaker praises Kubernetes as a tool for efficient cloud operations and carbon reduction, highlighting its role in achieving high container utilization rates and resource optimization. This keyword exemplifies the practical solutions that can be implemented in the tech industry to support sustainability goals.

๐Ÿ’กVertical Pod Autoscaler

Vertical Pod Autoscaler (VPA) is a Kubernetes feature that automatically adjusts the resources allocated to a pod based on its CPU and memory usage. The speaker mentions VPA as a tool that empowers developers to optimize their applications' resource usage, leading to more efficient operations and reduced energy consumption. This keyword is an example of the specific technical tools that can be leveraged to address the broader issue of sustainability in software development and operations.

๐Ÿ’กGreen Digitalization

Green digitalization refers to the process of implementing digital technologies in an environmentally sustainable way, aiming to reduce the carbon footprint of digital processes. The speaker introduces the concept of 'green digitalization' as a strategic initiative to address the energy consumption associated with digitalization, emphasizing the need for a sustainable approach to technology development and use.

๐Ÿ’กCarbon Compensation

Carbon compensation is a strategy to offset carbon dioxide emissions by investing in projects that reduce or remove an equivalent amount of CO2 elsewhere. The speaker critiques the reliance on carbon compensation, arguing that it should not be seen as a solution to the exponential growth in energy demands, but rather as one part of a broader strategy to address climate change. This keyword is important in understanding the complexities of climate action and the need for a multifaceted approach.

๐Ÿ’กDeveloper Empowerment

Developer empowerment refers to the process of providing developers with the tools, knowledge, and autonomy to make decisions that positively impact the environment and efficiency of software development. The speaker emphasizes the role of developers in driving change towards sustainability through their everyday decisions and actions. This keyword highlights the importance of engaging and enabling the tech community to lead by example in sustainable practices.

๐Ÿ’กEnergy Metrics

Energy metrics are quantitative measures used to assess and track energy consumption and efficiency. In the video, the speaker discusses the importance of having access to energy metrics to understand the impact of workloads and to optimize resource usage. This keyword is central to the argument that transparency and data-driven insights are crucial for making informed decisions that support sustainability goals.

Highlights

The speaker shares a personal struggle with flu and uses it as a metaphor for needing help to create a 'wave' of action.

Growing up in an ugly suburb sparked the speaker's appreciation for nature through surfing, highlighting the importance of personal experiences in environmental awareness.

The urgency of climate change is emphasized, noting that environmental issues were recognized in the '90s but political action was insufficient.

The speaker's involvement with a company focused on sustainable mobility through train travel, underscoring the role of individual choices in climate action.

The exponential growth in electricity consumption due to digitalization is discussed, with a prediction of needing the equivalent of Germany's consumption by 2025.

The limitations of current cloud provider tools in measuring application-specific carbon emissions are highlighted, indicating a need for better metrics.

The concept of 'green digitalization' is introduced, showing the industry's response to the environmental impact of technology.

Kubernetes is praised as a tool for efficient cloud operations and carbon reduction, demonstrating the potential of technology to address environmental issues.

The importance of developer empowerment in achieving climate targets is stressed, as they are the decision-makers in software development.

The use of vertical pod autoscaling is suggested to optimize resource usage, reducing waste and improving efficiency.

The concept of scheduling workloads to align with human work patterns is proposed to save energy, such as using down scalers to turn off environments during off-hours.

Information visibility is deemed crucial for making informed decisions on energy usage, with the introduction of a tool like KLA for energy metrics.

The impact of developer actions on carbon emissions is discussed, with the idea of correlating code changes with energy consumption.

The importance of starting now and taking small steps towards energy efficiency and green technology is emphasized.

The call to focus on empowering developers and leveraging their intrinsic motivation to care for the planet is highlighted.

Transcripts

play00:00

hello Paris hello

play00:01

everyone I hope you are having a great

play00:03

time are you having a great

play00:08

time that's good to know because I'm

play00:10

having a terrible

play00:12

time I caught this flu this weekend I

play00:16

thought how I will I stand to this Stage

play00:18

15 minutes talking

play00:20

to 180 M

play00:22

all and tonight I could not sleep with

play00:25

fever and I had this kind of

play00:27

hallucination dreams and so on you know

play00:29

and I got an idea there

play00:32

and I I I will need your help so I ask

play00:35

you that you if you have things on your

play00:38

lap and so on put them on a safe place

play00:40

under your

play00:41

chair and

play00:45

um first I want to talk why I I I need

play00:49

this I I grew up in a kind of ugly

play00:52

suburb of an very beautiful capital city

play00:55

at the Atlantic and my first Contact to

play00:57

Nature and uh my interaction with nature

play01:01

was when going to to the coast to Surf

play01:04

and you have this moments when you are

play01:07

surfing and you can be completely wasted

play01:10

either because you partti the last days

play01:12

or because you were surfing the whole

play01:15

day but when there's this wave coming

play01:18

you get this adrenaline and this energy

play01:20

because you want to catch this wave and

play01:22

you definitely don't want to be caught

play01:23

by the wave so what I'm going to ask you

play01:26

is that we make a big wave and show the

play01:29

power of the cuon in Paris and for this

play01:33

I we I will give you a signal on three

play01:36

and the of course the first sles go and

play01:38

you know how wave Works no so let's go

play01:41

one two

play01:44

three

play01:47

ah this was not the wave I would like to

play01:49

catch come on I want I I want energy I

play01:52

need your energy and your impulse and I

play01:54

also want to hear the the the Overflow

play01:57

rooms and the people at home so everyone

play02:00

needs to make so again one 2 3

play02:05

go up to the back yeah thank you thank

play02:09

you very much so this is the start of

play02:13

the

play02:16

[Music]

play02:19

story the story I'm going to tell you is

play02:22

precisely about the urgency of climate

play02:25

change already in the '90s when I was

play02:27

surfing we had already all this

play02:29

environment issues coming up and um but

play02:33

the politics were still not so so far

play02:36

with it we had the first meetings in in

play02:39

uh the Rio Janeiro in '92 we had the

play02:41

first Kyoto Protocol with the first

play02:43

commitments to climate targets but they

play02:47

were all kind of insufficient although

play02:49

science already knew that it's a urgent

play02:51

matter and requires a multi-level action

play02:55

and engaging every individual and

play02:57

organization to achieve the targets that

play02:59

we need

play03:01

and um it was not difficult when a

play03:03

friend of mine um contact me uh one day

play03:07

and and said hey we have this team at

play03:09

the Deutsche bar and we do kubernetes

play03:12

and so on and yeah I love I love this

play03:15

and but I also said yeah this is

play03:17

actually a um company with a purpose

play03:20

because it's I Love Trains I love train

play03:22

travel and um I like to

play03:26

um I think sustainable Mobility is um

play03:30

with trains is a very important to

play03:32

achieve the climate targets and our

play03:34

company also believes this so we are

play03:36

pushing forward to more Train Transport

play03:38

for Freight and passengers and for

play03:41

achieving this we obvious cannot just

play03:44

build infinitely more infrastructure

play03:45

this is quite heavy infrastructure we

play03:47

also need to optimize the infrastructure

play03:49

make it more efficient more modern and

play03:51

this requires a highly degree of

play03:53

digitalization but digitalization comes

play03:56

with an with an

play03:57

impact and the impact is according to

play04:00

the International Energy agency quite uh

play04:03

remarkable we see that in the next three

play04:05

years according to the last report we

play04:07

will at least require the amount of

play04:10

electricity consumption of Sweden and in

play04:13

the worst case scenario the electricity

play04:16

consumption of Germany so we are

play04:18

reaching an exponential consumption

play04:20

phase and of

play04:22

course there our um approaches to reduce

play04:26

it but we also want to approach that the

play04:31

deutan started a green digitalization

play04:33

initiative in

play04:35

20222 to address this

play04:37

question and it came with the support of

play04:41

CEOs and cios but it immediately raised

play04:44

also a very strong Grassroots movement

play04:46

from developers from employers of the

play04:49

company that started raising a lot of

play04:51

initiatives to um small initiatives on

play04:54

their free times uh making small

play04:57

contributions uh to measure for example

play05:00

um the the emissions of websites or uh

play05:03

to make some papers on how to uh reduce

play05:05

the impact on your home office work all

play05:08

kinds of initiatives raising there

play05:12

and we also had a project where we

play05:14

wanted to measure and to bring

play05:16

developers to be able to measure the

play05:18

impact of our workloads and we started

play05:21

because we are a cloud uh based uh we

play05:23

have a cloud cloud-based approach uh we

play05:26

started to look at what the cloud

play05:28

providers

play05:30

uh give us in terms of tools and

play05:33

unfortunately we did not find uh very um

play05:36

helpful information because the tools um

play05:39

have

play05:40

limitations they are not able to assign

play05:43

to an application so if I have on my

play05:45

accounts multiple

play05:47

applications uh there's no tax support

play05:50

uh I cannot distinguish it uh the

play05:52

granularity is very low so I have 50%

play05:55

ec2 and um 20% everything else it's not

play05:58

telling me much

play06:00

it's not very actual sometimes you get

play06:02

the information three months after uh

play06:04

the data has been uh created and it does

play06:08

not give you any information on the

play06:09

energy use why is this important because

play06:13

emissions there are calculated also with

play06:15

compensation and this means that you are

play06:18

not able to establish a casual casual

play06:20

relationship between the actions that

play06:22

you do on your all day and um the the

play06:25

effect on

play06:27

emissions and to understand also the

play06:29

problem of compensation uh I would like

play06:31

to ask who is your fan of Doctor Who

play06:35

series yeah quite some yeah great

play06:38

science fiction series I watch it with

play06:40

my daughters every uh evening almost and

play06:43

I we hope still to finish it before they

play06:46

leave home but um yeah there's this

play06:49

episode where they say suddenly wake up

play06:51

in middle of London and London is full

play06:53

of a forest so this is for me a good

play06:56

metaphor of compensation of course

play06:58

compensating carbon is an important

play07:00

measure but we cannot think that with an

play07:02

exponential growth scenario of

play07:05

exponential growth of energy demands

play07:07

that we will be able to compensate all

play07:09

this energy because there are

play07:10

effectively limits to growth physical

play07:12

limits and and so we need to address the

play07:16

different strategies for um sufficiency

play07:21

what I really need and why for

play07:23

efficiency how do I minimize the

play07:25

resource used at produ and production

play07:28

and consistent

play07:30

which would be the case for example of

play07:32

of compensation so focusing first of all

play07:35

on the efficient part and on the role of

play07:38

developers back to the agents on this

play07:41

talk I had the to topic developer

play07:43

empowerment so developers are

play07:45

effectively effectively the everyday

play07:47

decision makers in what comes to

play07:50

software we don't need management papers

play07:53

to um to say we we are doing this and

play07:56

achieving this targets if we don't

play07:58

engage Developers and give them the

play08:00

tools in the hand we will not be able to

play08:02

make a change in the way we develop code

play08:05

in the way we um manage infrastructure

play08:08

so we POS the question what tools and

play08:11

and approaches are able to empower these

play08:15

developers let's start with

play08:17

platforms we have a platform strategy at

play08:20

the Doan and in the last years we comb

play08:23

converge uh the whole subsidiaries to um

play08:27

enforce standardization in this in this

play08:30

um level and the importance of platforms

play08:32

is uh on this aspect is that we can

play08:36

leverage effects and provide an high

play08:39

level of standardization and open to

play08:41

contributions that improve it

play08:43

continuously and provide safe same

play08:46

design um aspects that everyone can use

play08:49

without even thinking am I doing it

play08:51

green or not it's just um helping you to

play08:54

be green by

play08:56

default I myself I'm a bit biased I love

play08:59

kubernetes I was product owner of two

play09:00

teams uh in this field and uh have a

play09:04

personal affility to this but I believe

play09:06

kubernetes is a tool for efficient Cloud

play09:09

operations and for carbon F reduction

play09:12

per Excellence it's not only a container

play09:14

orchestration tool it's much more

play09:16

powerful than that it's a platform

play09:18

building tool and it's a green it

play09:20

tool we have uh with our clusters

play09:24

achieved one of the um interesting uh

play09:28

very high density of containers we have

play09:30

some shared clusters and we had uh then

play09:34

put of course the node AO scaling and we

play09:37

managed to achieve around 70% of CPU

play09:40

utilization on these shared clusters

play09:43

it's quite a high utilization so you

play09:45

would say well we are effective we don't

play09:46

need to do anything

play09:48

else the fact is however only 15% of

play09:51

this utilization was effectively used by

play09:54

the applications we could see for this

play09:56

for example on graph dashboards where

play09:59

there are a lot of reserved um CPU

play10:02

utilization but only a very small part

play10:05

of it is being used why is this happen

play10:09

well once it runs no one cares we are we

play10:12

have a lot of things to do in our

play10:14

business all day so I Define requests

play10:16

and

play10:18

limits um traditionally um the API uh um

play10:22

request limits 512 megabytes it is

play10:25

running I make it conservative because I

play10:27

think maybe there will be a lot of

play10:29

people coming and then I cannot burst so

play10:32

rather have it conservative and it stays

play10:35

there and it's taking CPU it's occupying

play10:38

the node and wasting

play10:41

resources fortunately kubernetes offers

play10:43

us a vertical P AO scalar there also of

play10:46

course horizontal Part auto scal and so

play10:48

on but I find this one particularly

play10:49

interesting because it is taking out of

play10:52

the ends of the developer they need to

play10:55

think what would be the appropriate

play10:57

resources and rather making

play10:59

recommendations or even automatically

play11:01

adjusting the container workloads to be

play11:04

optimized to their

play11:05

needs so let's go out with expensive

play11:08

getwork and make sure to use vertical

play11:11

pod outo scalar the second aspect

play11:14

related to the sufficient see is

play11:18

scheduling and kubernetes is also good

play11:20

on this you know we have all these

play11:22

worker rights movements that manage that

play11:24

we are able to work eight hours a day

play11:26

and have eight hours to do whatever we

play11:28

want and eight hours to sleep

play11:30

I don't think most of you do this anyway

play11:32

but but let's assume everyone works from

play11:34

9 to 5 like like like would be the

play11:37

traditional approach this means two3 of

play11:41

the day there's no one at the office

play11:44

and your workloads are running there

play11:48

development environments test

play11:50

environments we testing it with test

play11:53

testing stuff at midnight just turn it

play11:56

off use Cube down scaler annotate your

play11:59

deployments so so simple like that it

play12:02

will turn off when you leave the office

play12:04

it will turn on again when you when you

play12:07

come back and it's working you don't

play12:09

even need to care for it it does it

play12:10

every day it saves a lot of energy and

play12:12

it costs

play12:14

nothing last

play12:16

aspect we need information

play12:19

visibility it's like if I am traveling

play12:22

by train and my train is late or it

play12:24

fails what do I do if I did not have

play12:26

good data informing me providing me how

play12:30

how will be my

play12:34

Alternatives so this is where we came

play12:36

with KLA we wanted to grab the energy

play12:40

metrics and I will not go into detail in

play12:42

Kepler here because there are 50 at

play12:45

least 50 people in this room for sure

play12:46

that know in detail how Kepler um

play12:49

evaluates um energy statistics and

play12:52

transforms this into uh into wats uh

play12:57

but the interesting there is that you

play13:00

can

play13:01

actually get information from high level

play13:04

up to the container level so really have

play13:08

information on real on near real time on

play13:10

how much energy your single components

play13:13

are using and you know in grafana it's

play13:16

also possible you can make for example

play13:18

an annotation from your pipeline and

play13:20

Associate a certain change in your code

play13:24

with an impact on the

play13:27

consumption with we tried to roll it out

play13:29

we thought that's actually yeah great

play13:31

open source it's available let's just

play13:33

roll it out it was not so easy the first

play13:35

one was the Enterprise Readiness in

play13:37

terms of security there were some

play13:39

unneeded dependencies um there that we

play13:42

needed to

play13:43

remove it were able to give give a

play13:45

contribution which was

play13:47

great and the second one was well there

play13:51

were a lot of metrics so there was

play13:53

someone being called in the night on the

play13:55

on call duty and you know how people

play13:57

hate this no one like to be called in

play13:59

the night for because the monitoring

play14:01

system is overloaded so we had to reduce

play14:04

the um scraping interval which was by

play14:08

default on 3 seconds to much less and

play14:11

actually we don't want to keep also so

play14:12

much data so it's it's enough if we have

play14:15

10 20 seconds scraping to give this

play14:17

information about uh what is the effect

play14:20

of the different

play14:22

commits so data is the source of

play14:25

everything and we have seen on the left

play14:28

side where you see the green is the

play14:30

kubernetes pushing back to grafana and

play14:33

it allowed us to make this beautiful

play14:36

dashboards dashboards are always

play14:37

something beautiful I I find them at at

play14:39

least and I think many people love

play14:42

dashboards and um with this we had the

play14:45

first tool that really brings this in

play14:47

the hand of developers to to make a

play14:50

change and then we of course started

play14:53

getting more and more people interested

play14:55

on this and to gather the whole

play14:59

Cloud carbon footprint um data and we

play15:03

required a middle layer with our data L

play15:06

with the data with a data set with data

play15:08

governance that is in the middle and

play15:10

allows for other systems to couple and

play15:13

also create information for other

play15:17

potential agents like for example

play15:19

architecture management and Enterprise

play15:21

architecture management we see here um

play15:24

an example of the dashboard that we are

play15:25

building with our Enterprise arit

play15:28

management system system and um so um

play15:32

it's about really bringing the tools

play15:34

also to and and the information to the

play15:36

tools that are already being used by

play15:39

these people so as a final remarks I

play15:42

would like to say the first important

play15:46

thing is start start now it is urgent

play15:49

the issue is urgent and we are all in an

play15:51

early stage of knowing how to actually

play15:54

measure energy we have lack of

play15:56

information we don't know if this data

play15:57

is good but it's better as no data some

play16:00

information is better as no information

play16:02

so just start you will see how many

play16:05

people because really developers have an

play16:08

intrinsic motivation not really to save

play16:10

costs but everyone has an intrinsic

play16:12

motivation to care for our planet I

play16:14

believe it so just start small and

play16:17

beautiful and you will see how your

play16:19

community and your e system around the

play16:21

platforms and around green it will um

play16:25

grow and focus on empowerment focus on

play16:28

engine on the agents on the people that

play16:31

are on the all day taking decisions with

play16:34

their actions encourage developer

play16:36

initiative and their intrisic

play16:39

motivation so thank you very much um

play16:42

let's stick together if you want contact

play16:44

me or any of my colleagues that are also

play16:47

on this uh conference and um it was a

play16:51

pleasure to be here and I would like to

play16:53

to ask the moderator to come here and

play16:57

because I want you also to experience

play16:58

this wave yeah so let's do it one more

play17:01

time if you don't mind H so one two

play17:07

3 thank you very much have a nice day

play17:10

thank you so much

play17:11

[Music]

play17:13

Quon

Rate This
โ˜…
โ˜…
โ˜…
โ˜…
โ˜…

5.0 / 5 (0 votes)

Related Tags
Climate ChangeDigital InnovationSustainable TechIndividual EmpowermentTech DevelopmentGreen InitiativesKubernetesEnergy EfficiencyDeveloper MotivationCorporate Responsibility