Up next in 10
Join this live session with Magnus Mårtensson ft. Jennifer Ritzinger for the next episode of The Cloud Show with Magnus Mårtensson on August 21, at 01:05 PM (EST).
The show is about cloud leadership and all the important questions relating to cloud projects. Certainly, many matters when a company is going to and wants to be successful in the cloud, are about technology. However, there are many additional matters, adjacent to technology, that we also need to tend to regarding business strategy, human resources, organizational change, planning for a technical cloud approach, and many more questions. These conversations are critical for a healthy cloud and for a swift and accurate cloud approach.
📺 CSharp TV - Dev Streaming Destination http://csharp.tv
🌎 C# Corner - Community of Software and Data Developers https://www.c-sharpcorner.com
#CSharpTV #csharpcorner #TheCloudShow #CSharp #Interview
Show More Show Less View Video Transcript
0:04
hello everyone welcome back to another
0:06
hello everyone welcome back to another
0:06
hello everyone welcome back to another episode of the cloud show I have an
0:09
episode of the cloud show I have an
0:09
episode of the cloud show I have an awesome guest today he has maybe the
0:12
awesome guest today he has maybe the
0:12
awesome guest today he has maybe the best job a tech person can have in the
0:14
best job a tech person can have in the
0:14
best job a tech person can have in the world which is amazing we're going to be
0:17
world which is amazing we're going to be
0:17
world which is amazing we're going to be talking about Cloud strategy today we're
0:20
talking about Cloud strategy today we're
0:20
talking about Cloud strategy today we're going to talk about how to move from
0:21
going to talk about how to move from
0:21
going to talk about how to move from having a cloud first strategy whatever
0:24
having a cloud first strategy whatever
0:24
having a cloud first strategy whatever that means into moving towards becoming
0:27
that means into moving towards becoming
0:27
that means into moving towards becoming really truly Cloud native and the guest
0:30
really truly Cloud native and the guest
0:30
really truly Cloud native and the guest today he works at LEGO Group which is
0:32
today he works at LEGO Group which is
0:32
today he works at LEGO Group which is amazing he works for Lego come on and
0:35
amazing he works for Lego come on and
0:35
amazing he works for Lego come on and his name is Rasmus
0:38
his name is Rasmus
0:38
his name is Rasmus [Music]
0:48
Hal hello Rasmus my friend hi Magnus my
0:52
Hal hello Rasmus my friend hi Magnus my
0:52
Hal hello Rasmus my friend hi Magnus my friend good to see you the same it's
0:54
friend good to see you the same it's
0:54
friend good to see you the same it's good to see you too you know everyone in
0:56
good to see you too you know everyone in
0:56
good to see you too you know everyone in Tech hates you
0:58
Tech hates you
0:58
Tech hates you right why because you work for Lego man
1:01
right why because you work for Lego man
1:01
right why because you work for Lego man it's like we want to kill you and take
1:03
it's like we want to kill you and take
1:03
it's like we want to kill you and take over your life ah okay okay if fa then I
1:08
over your life ah okay okay if fa then I
1:08
over your life ah okay okay if fa then I put it down freely I'm not gonna deny it
1:10
put it down freely I'm not gonna deny it
1:10
put it down freely I'm not gonna deny it I know I know it's envious to work for
1:12
I know I know it's envious to work for
1:12
I know I know it's envious to work for Lego yeah it's
1:14
Lego yeah it's
1:14
Lego yeah it's amazing okay but I will tell you um it's
1:18
amazing okay but I will tell you um it's
1:18
amazing okay but I will tell you um it's wonderful we play a lot right we have a
1:20
wonderful we play a lot right we have a
1:20
wonderful we play a lot right we have a lot of fun uh actually just uh Friday
1:23
lot of fun uh actually just uh Friday
1:23
lot of fun uh actually just uh Friday last week we had a international uh play
1:26
last week we had a international uh play
1:26
last week we had a international uh play day so it was all our sites all over the
1:28
day so it was all our sites all over the
1:28
day so it was all our sites all over the world took the Friday off
1:30
world took the Friday off
1:30
world took the Friday off to go and play with the colleagues right
1:32
to go and play with the colleagues right
1:32
to go and play with the colleagues right it was amazing it was Festival that's
1:35
it was amazing it was Festival that's
1:35
it was amazing it was Festival that's amazing yeah that's great okay well the
1:38
amazing yeah that's great okay well the
1:39
amazing yeah that's great okay well the downside Magnus and everyone is that you
1:41
downside Magnus and everyone is that you
1:41
downside Magnus and everyone is that you don't I mean it's really difficult not
1:45
don't I mean it's really difficult not
1:45
don't I mean it's really difficult not to go and and and buy some stuff right
1:48
to go and and and buy some stuff right
1:48
to go and and and buy some stuff right so uh I recently acquired the The
1:50
so uh I recently acquired the The
1:50
so uh I recently acquired the The Dungeons and dragon
1:53
Dungeons and dragon
1:53
Dungeons and dragon set to pick it up so and then you have
1:55
set to pick it up so and then you have
1:55
set to pick it up so and then you have to have time to build all the sets you
1:57
to have time to build all the sets you
1:57
to have time to build all the sets you know that's it's a big job
2:00
know that's it's a big job
2:00
know that's it's a big job I know right ah terrible all right so
2:03
I know right ah terrible all right so
2:03
I know right ah terrible all right so you work for Lego and uh let's see if I
2:06
you work for Lego and uh let's see if I
2:06
you work for Lego and uh let's see if I can recall you're basically are like
2:09
can recall you're basically are like
2:09
can recall you're basically are like developer what's what's your title again
2:11
developer what's what's your title again
2:11
developer what's what's your title again sorry oh it's a fancy title um that
2:14
sorry oh it's a fancy title um that
2:14
sorry oh it's a fancy title um that comes with the job so I'm head of
2:17
comes with the job so I'm head of
2:17
comes with the job so I'm head of developer experience and that's what I
2:20
developer experience and that's what I
2:20
developer experience and that's what I was out Foundation um basically it's
2:23
was out Foundation um basically it's
2:23
was out Foundation um basically it's everything with hosting so majority of
2:25
everything with hosting so majority of
2:25
everything with hosting so majority of that is cloud for us but we do have
2:28
that is cloud for us but we do have
2:28
that is cloud for us but we do have factories where we have AG
2:31
factories where we have AG
2:31
factories where we have AG Computing is where we do hosting right
2:33
Computing is where we do hosting right
2:33
Computing is where we do hosting right and then I do developer experience so
2:36
and then I do developer experience so
2:36
and then I do developer experience so pretty much whatever can make our
2:38
pretty much whatever can make our
2:38
pretty much whatever can make our developers engaged more productive more
2:41
developers engaged more productive more
2:41
developers engaged more productive more happy that's brilliant what an awesome
2:44
happy that's brilliant what an awesome
2:44
happy that's brilliant what an awesome job all right so obviously you came to
2:47
job all right so obviously you came to
2:47
job all right so obviously you came to Lego to to work with their Cloud
2:49
Lego to to work with their Cloud
2:50
Lego to to work with their Cloud strategy and help Lego with Cloud you
2:52
strategy and help Lego with Cloud you
2:52
strategy and help Lego with Cloud you had previous long Cloud experience and
2:54
had previous long Cloud experience and
2:55
had previous long Cloud experience and that's why you came to Lego um and so
2:59
that's why you came to Lego um and so
2:59
that's why you came to Lego um and so you're talking about uh strategy wise um
3:03
you're talking about uh strategy wise um
3:03
you're talking about uh strategy wise um how to have a cloud first strategy many
3:07
how to have a cloud first strategy many
3:07
how to have a cloud first strategy many companies have Define that they should
3:08
companies have Define that they should
3:08
companies have Define that they should have we shall have a cloud first
3:10
have we shall have a cloud first
3:10
have we shall have a cloud first strategy and then that's what's the
3:13
strategy and then that's what's the
3:13
strategy and then that's what's the difference between having between saying
3:15
difference between having between saying
3:15
difference between having between saying that having that on a piece of paper
3:16
that having that on a piece of paper
3:16
that having that on a piece of paper that says we have a cloud first strategy
3:18
that says we have a cloud first strategy
3:18
that says we have a cloud first strategy and the difference between that and
3:20
and the difference between that and
3:20
and the difference between that and being truly Cloud native yeah yeah
3:23
being truly Cloud native yeah yeah
3:23
being truly Cloud native yeah yeah probably some of uh the the big
3:26
probably some of uh the the big
3:26
probably some of uh the the big Management Consultant firms have uh
3:28
Management Consultant firms have uh
3:28
Management Consultant firms have uh aligned their thinking and
3:30
aligned their thinking and
3:30
aligned their thinking and made PowerPoint saying that any
3:32
made PowerPoint saying that any
3:32
made PowerPoint saying that any Enterprise should have a cloud first
3:33
Enterprise should have a cloud first
3:33
Enterprise should have a cloud first strategy because it's quite common right
3:35
strategy because it's quite common right
3:35
strategy because it's quite common right yeah but I really think we need to to
3:38
yeah but I really think we need to to
3:38
yeah but I really think we need to to depict that um so what I did when I
3:41
depict that um so what I did when I
3:41
depict that um so what I did when I joined the group uh three years ago was
3:44
joined the group uh three years ago was
3:44
joined the group uh three years ago was I draw drew up a a new Cloud strategy
3:48
I draw drew up a a new Cloud strategy
3:48
I draw drew up a a new Cloud strategy for H Lego group that outlined we were
3:51
for H Lego group that outlined we were
3:51
for H Lego group that outlined we were moving towards Cloud first um and what
3:56
moving towards Cloud first um and what
3:56
moving towards Cloud first um and what we Lally have been doing is refreshing
3:58
we Lally have been doing is refreshing
3:58
we Lally have been doing is refreshing that strategy and now we are abandoning
4:02
that strategy and now we are abandoning
4:02
that strategy and now we are abandoning the cloud first I'll get back to it why
4:05
the cloud first I'll get back to it why
4:05
the cloud first I'll get back to it why uh but basically saying we want to build
4:07
uh but basically saying we want to build
4:08
uh but basically saying we want to build not Cloud first but Cloud native and I
4:10
not Cloud first but Cloud native and I
4:10
not Cloud first but Cloud native and I would like to unpack that as well yeah
4:13
would like to unpack that as well yeah
4:13
would like to unpack that as well yeah yeah definitely what is the difference
4:15
yeah definitely what is the difference
4:15
yeah definitely what is the difference for for people who don't know I'm sure a
4:17
for for people who don't know I'm sure a
4:17
for for people who don't know I'm sure a lot of you know managers administrators
4:19
lot of you know managers administrators
4:19
lot of you know managers administrators out there they've heard about this
4:21
out there they've heard about this
4:21
out there they've heard about this having this fancy Cloud first strategy
4:23
having this fancy Cloud first strategy
4:23
having this fancy Cloud first strategy but then what has that actually really
4:25
but then what has that actually really
4:25
but then what has that actually really really entail yeah exactly and I I think
4:29
really entail yeah exactly and I I think
4:30
really entail yeah exactly and I I think the for me at least and what what we
4:32
the for me at least and what what we
4:32
the for me at least and what what we have been steering towards with the
4:34
have been steering towards with the
4:34
have been steering towards with the cloud first strategy was just saying
4:36
cloud first strategy was just saying
4:36
cloud first strategy was just saying that what cloud is our first choice for
4:39
that what cloud is our first choice for
4:39
that what cloud is our first choice for hosting applications Y and we would even
4:42
hosting applications Y and we would even
4:42
hosting applications Y and we would even unpack that a little further and say we
4:44
unpack that a little further and say we
4:45
unpack that a little further and say we actually prefer hosting with software as
4:47
actually prefer hosting with software as
4:47
actually prefer hosting with software as a service uh for anything that's
4:49
a service uh for anything that's
4:49
a service uh for anything that's nondifferentiating so rather than
4:51
nondifferentiating so rather than
4:51
nondifferentiating so rather than running your HR System you would go and
4:53
running your HR System you would go and
4:53
running your HR System you would go and buy uh a software as a service uh
4:57
buy uh a software as a service uh
4:57
buy uh a software as a service uh application that did that for the
4:58
application that did that for the
4:58
application that did that for the Enterprise
5:00
Enterprise
5:00
Enterprise uh so that what the preferred uh way of
5:02
uh so that what the preferred uh way of
5:02
uh so that what the preferred uh way of Hosting but when we have differentiating
5:05
Hosting but when we have differentiating
5:05
Hosting but when we have differentiating workloads of course where we do custom
5:07
workloads of course where we do custom
5:07
workloads of course where we do custom development that's when we use the cloud
5:09
development that's when we use the cloud
5:09
development that's when we use the cloud providers um out there and there we
5:13
providers um out there and there we
5:13
providers um out there and there we would say it's Cloud first right now in
5:16
would say it's Cloud first right now in
5:16
would say it's Cloud first right now in the last three years what we've been
5:17
the last three years what we've been
5:17
the last three years what we've been doing is also taking some of the Legacy
5:20
doing is also taking some of the Legacy
5:20
doing is also taking some of the Legacy workloads or the classic workloads
5:22
workloads or the classic workloads
5:22
workloads or the classic workloads because we had a data center that did
5:24
because we had a data center that did
5:24
because we had a data center that did General compute hosting uh and then
5:27
General compute hosting uh and then
5:27
General compute hosting uh and then pushing that towards Cloud so just
5:30
pushing that towards Cloud so just
5:30
pushing that towards Cloud so just getting rid of the rest you could say
5:32
getting rid of the rest you could say
5:32
getting rid of the rest you could say and doing that bit of spring cleaning uh
5:34
and doing that bit of spring cleaning uh
5:34
and doing that bit of spring cleaning uh just to make sure that uh that critical
5:38
just to make sure that uh that critical
5:38
just to make sure that uh that critical last piece um you get rid of that or get
5:41
last piece um you get rid of that or get
5:41
last piece um you get rid of that or get that to Cloud right y I think that is
5:44
that to Cloud right y I think that is
5:44
that to Cloud right y I think that is probably the key to our success is we
5:46
probably the key to our success is we
5:46
probably the key to our success is we have had a transition over three five
5:49
have had a transition over three five
5:49
have had a transition over three five years with moving to software as a
5:50
years with moving to software as a
5:51
years with moving to software as a service yeah using Cloud as the number
5:54
service yeah using Cloud as the number
5:54
service yeah using Cloud as the number the preferred option yeah and ending
5:56
the preferred option yeah and ending
5:56
the preferred option yeah and ending that strategy cycle with moving the rest
5:59
that strategy cycle with moving the rest
5:59
that strategy cycle with moving the rest uh as a sort of a liftt and shift
6:02
uh as a sort of a liftt and shift
6:02
uh as a sort of a liftt and shift approach and then we modernize when we
6:04
approach and then we modernize when we
6:04
approach and then we modernize when we are in Cloud so right that's where we're
6:07
are in Cloud so right that's where we're
6:07
are in Cloud so right that's where we're coming from cloud first means software
6:09
coming from cloud first means software
6:09
coming from cloud first means software as a service it means putting whatever
6:12
as a service it means putting whatever
6:12
as a service it means putting whatever you can in Cloud but for us factories
6:15
you can in Cloud but for us factories
6:15
you can in Cloud but for us factories are all over the world right and they
6:17
are all over the world right and they
6:17
are all over the world right and they are in remote area and latency matters
6:20
are in remote area and latency matters
6:20
are in remote area and latency matters so there we will have Edge Computing
6:22
so there we will have Edge Computing
6:22
so there we will have Edge Computing sites right simply have uh like data set
6:25
sites right simply have uh like data set
6:25
sites right simply have uh like data set setups but exclusively for that site
6:27
setups but exclusively for that site
6:28
setups but exclusively for that site yeah so you have you have the site
6:30
yeah so you have you have the site
6:30
yeah so you have you have the site specific or site located um Edge compute
6:35
specific or site located um Edge compute
6:35
specific or site located um Edge compute scenario those those will exist because
6:37
scenario those those will exist because
6:37
scenario those those will exist because you need them you like like you said
6:39
you need them you like like you said
6:39
you need them you like like you said latency is a thing um and and for the
6:42
latency is a thing um and and for the
6:42
latency is a thing um and and for the rest for everything else like the the
6:44
rest for everything else like the the
6:44
rest for everything else like the the the the purchasing systems and sales and
6:47
the the purchasing systems and sales and
6:47
the the purchasing systems and sales and and I guess um Logistics and right all
6:52
and I guess um Logistics and right all
6:52
and I guess um Logistics and right all all all the other things that are not
6:55
all all the other things that are not
6:55
all all the other things that are not manufacturing you have the the real
6:58
manufacturing you have the the real
6:58
manufacturing you have the the real cloud nativeness have you uh completed a
7:02
cloud nativeness have you uh completed a
7:02
cloud nativeness have you uh completed a transition away from having any other
7:04
transition away from having any other
7:04
transition away from having any other data center than Cloud yet or that's
7:07
data center than Cloud yet or that's
7:07
data center than Cloud yet or that's actually why we now uh actually uh this
7:10
actually why we now uh actually uh this
7:10
actually why we now uh actually uh this week we are announcing our refreshed
7:13
week we are announcing our refreshed
7:13
week we are announcing our refreshed Cloud strategy and now it's called a
7:15
Cloud strategy and now it's called a
7:15
Cloud strategy and now it's called a cloud an edge hosting strategy oh that's
7:19
cloud an edge hosting strategy oh that's
7:19
cloud an edge hosting strategy oh that's cool compass that we will be providing
7:21
cool compass that we will be providing
7:21
cool compass that we will be providing compute anywhere it's needed but cloud
7:25
compute anywhere it's needed but cloud
7:25
compute anywhere it's needed but cloud is our our uh that's our center right
7:28
is our our uh that's our center right
7:28
is our our uh that's our center right that that's our core
7:30
that that's our core
7:30
that that's our core and then we call out specifically a
7:33
and then we call out specifically a
7:33
and then we call out specifically a transition from cloud first because I
7:35
transition from cloud first because I
7:35
transition from cloud first because I believe everything is hosted where it
7:37
believe everything is hosted where it
7:37
believe everything is hosted where it should be so take that with an asteris
7:40
should be so take that with an asteris
7:40
should be so take that with an asteris right but that's for majority of our
7:43
right but that's for majority of our
7:43
right but that's for majority of our workloads they are hosted where they
7:44
workloads they are hosted where they
7:44
workloads they are hosted where they should be right so we are in a good
7:47
should be right so we are in a good
7:47
should be right so we are in a good place we don't need to push Cloud first
7:49
place we don't need to push Cloud first
7:49
place we don't need to push Cloud first anymore so we are transitioning into a
7:52
anymore so we are transitioning into a
7:52
anymore so we are transitioning into a Target state that talks about not where
7:55
Target state that talks about not where
7:55
Target state that talks about not where things are hosted so Cloud not Cloud but
7:58
things are hosted so Cloud not Cloud but
7:58
things are hosted so Cloud not Cloud but how they hosted yeah so addressing much
8:02
how they hosted yeah so addressing much
8:02
how they hosted yeah so addressing much more in the architecture level and this
8:04
more in the architecture level and this
8:04
more in the architecture level and this is where Cloud native comes in um and if
8:08
is where Cloud native comes in um and if
8:08
is where Cloud native comes in um and if we unpack that term a bit because it has
8:11
we unpack that term a bit because it has
8:11
we unpack that term a bit because it has been uh used by marketing it has been
8:15
been uh used by marketing it has been
8:15
been uh used by marketing it has been used by uh commercialized uh software so
8:19
used by uh commercialized uh software so
8:20
used by uh commercialized uh software so but if we actually look at what the
8:22
but if we actually look at what the
8:22
but if we actually look at what the cloud native Computing Foundation talks
8:25
cloud native Computing Foundation talks
8:25
cloud native Computing Foundation talks about uh when they talk about um
8:29
about uh when they talk about um
8:30
about uh when they talk about um being Cloud native it's not about
8:32
being Cloud native it's not about
8:32
being Cloud native it's not about kubernetes right it's actually about the
8:34
kubernetes right it's actually about the
8:34
kubernetes right it's actually about the characteristics of of the architecture
8:37
characteristics of of the architecture
8:37
characteristics of of the architecture yes it is there sort of uh there five
8:43
yes it is there sort of uh there five
8:43
yes it is there sort of uh there five characteristics that you need to to uh
8:46
characteristics that you need to to uh
8:46
characteristics that you need to to uh consider and called them out as this is
8:49
consider and called them out as this is
8:49
consider and called them out as this is the this is the way we build right so uh
8:51
the this is the way we build right so uh
8:51
the this is the way we build right so uh if we unpack the five the first one is
8:54
if we unpack the five the first one is
8:54
if we unpack the five the first one is Loosely coupled yep right Magnus you've
8:57
Loosely coupled yep right Magnus you've
8:57
Loosely coupled yep right Magnus you've been building with serverless for years
8:59
been building with serverless for years
8:59
been building with serverless for years right oh my God yes um and you have been
9:01
right oh my God yes um and you have been
9:01
right oh my God yes um and you have been building like you don't you don't start
9:03
building like you don't you don't start
9:03
building like you don't you don't start with the corporate Network right you
9:05
with the corporate Network right you
9:05
with the corporate Network right you don't start with uh creating firewalls
9:07
don't start with uh creating firewalls
9:07
don't start with uh creating firewalls and vpns and whatnot no you actually
9:09
and vpns and whatnot no you actually
9:10
and vpns and whatnot no you actually start by a small self-contained
9:13
start by a small self-contained
9:13
start by a small self-contained application uh that that um is is
9:16
application uh that that um is is
9:16
application uh that that um is is Loosely coupled from everything and then
9:19
Loosely coupled from everything and then
9:19
Loosely coupled from everything and then we use apis to integrate right we use
9:22
we use apis to integrate right we use
9:22
we use apis to integrate right we use apis to call the data we need we use
9:24
apis to call the data we need we use
9:24
apis to call the data we need we use apis and messaging to call the we need
9:28
apis and messaging to call the we need
9:28
apis and messaging to call the we need um and that way we we have Loosely
9:30
um and that way we we have Loosely
9:30
um and that way we we have Loosely coupled applications yeah definitely
9:33
coupled applications yeah definitely
9:33
coupled applications yeah definitely this massively allows us to secure our
9:36
this massively allows us to secure our
9:36
this massively allows us to secure our estate right because blast zones of
9:38
estate right because blast zones of
9:38
estate right because blast zones of individual application are much smaller
9:40
individual application are much smaller
9:40
individual application are much smaller but it also allows for a very much event
9:43
but it also allows for a very much event
9:43
but it also allows for a very much event driven architecture
9:45
driven architecture
9:45
driven architecture yeah I would I would agree that it go so
9:48
yeah I would I would agree that it go so
9:48
yeah I would I would agree that it go so far far as to say that it's it's sort of
9:50
far far as to say that it's it's sort of
9:50
far far as to say that it's it's sort of like an architectural style it's
9:53
like an architectural style it's
9:54
like an architectural style it's it's kind of right yes yeah um if at
9:59
it's kind of right yes yeah um if at
9:59
it's kind of right yes yeah um if at second characteristic it's sustainable
10:02
second characteristic it's sustainable
10:02
second characteristic it's sustainable and that's a word we we all care about
10:05
and that's a word we we all care about
10:05
and that's a word we we all care about of course uh yes but in in the in in the
10:08
of course uh yes but in in the in in the
10:08
of course uh yes but in in the in in the context of computing this is about
10:11
context of computing this is about
10:11
context of computing this is about efficient use of resources yes yes so
10:15
efficient use of resources yes yes so
10:15
efficient use of resources yes yes so where in the classic architecture you
10:17
where in the classic architecture you
10:17
where in the classic architecture you would start with a virtual machine which
10:20
would start with a virtual machine which
10:20
would start with a virtual machine which is a a fixed capacity model right you
10:22
is a a fixed capacity model right you
10:22
is a a fixed capacity model right you buy a
10:24
buy a
10:24
buy a capacity um even in Cloud right you you
10:26
capacity um even in Cloud right you you
10:27
capacity um even in Cloud right you you buy a certain VM that's it has a price
10:31
buy a certain VM that's it has a price
10:31
buy a certain VM that's it has a price tag and doesn't really matter if you use
10:32
tag and doesn't really matter if you use
10:32
tag and doesn't really matter if you use it or not you're paying for it right
10:35
it or not you're paying for it right
10:35
it or not you're paying for it right absolutely where when we talk about
10:36
absolutely where when we talk about
10:36
absolutely where when we talk about sustainable compute this is efficient
10:38
sustainable compute this is efficient
10:38
sustainable compute this is efficient use of resources and moving to that
10:40
use of resources and moving to that
10:40
use of resources and moving to that transaction based model of course we
10:43
transaction based model of course we
10:43
transaction based model of course we have to pay our providers right but uh
10:46
have to pay our providers right but uh
10:46
have to pay our providers right but uh but we can avoid over provisioning by
10:49
but we can avoid over provisioning by
10:49
but we can avoid over provisioning by having a transaction model and here I
10:51
having a transaction model and here I
10:51
having a transaction model and here I love serverless right so using AA
10:54
love serverless right so using AA
10:54
love serverless right so using AA functions Lambda functions uh because
10:57
functions Lambda functions uh because
10:57
functions Lambda functions uh because here you payer transaction rather than
11:00
here you payer transaction rather than
11:00
here you payer transaction rather than paying for the capacity and and your
11:02
paying for the capacity and and your
11:02
paying for the capacity and and your compute needs will just scale
11:05
compute needs will just scale
11:05
compute needs will just scale automatically with the needs of the
11:06
automatically with the needs of the
11:06
automatically with the needs of the business so I will give you an example
11:09
business so I will give you an example
11:09
business so I will give you an example lego.com is fairly
11:11
lego.com is fairly
11:11
lego.com is fairly popular um and when we release new stuff
11:15
popular um and when we release new stuff
11:15
popular um and when we release new stuff uh also when we do our Black Friday
11:17
uh also when we do our Black Friday
11:17
uh also when we do our Black Friday sales and our uh Insider
11:21
sales and our uh Insider
11:21
sales and our uh Insider weekends and our workloads uh are based
11:25
weekends and our workloads uh are based
11:25
weekends and our workloads uh are based on serverless right so they scale with
11:28
on serverless right so they scale with
11:28
on serverless right so they scale with us uh and that's just an amazing
11:30
us uh and that's just an amazing
11:30
us uh and that's just an amazing characteristic and super sustainable in
11:33
characteristic and super sustainable in
11:33
characteristic and super sustainable in in many way absolutely
11:36
in many way absolutely
11:36
in many way absolutely agreed the third one you're ready for it
11:40
agreed the third one you're ready for it
11:40
agreed the third one you're ready for it yes I'm ready for it
11:42
yes I'm ready for it
11:42
yes I'm ready for it manageable ah um and this really comes
11:45
manageable ah um and this really comes
11:45
manageable ah um and this really comes down to automation right yes yes it does
11:48
down to automation right yes yes it does
11:48
down to automation right yes yes it does so
11:49
so
11:49
so uh I mean uh one uh one devops famous
11:54
uh I mean uh one uh one devops famous
11:54
uh I mean uh one uh one devops famous person one said uh friends don't let
11:56
person one said uh friends don't let
11:56
person one said uh friends don't let friends right click deploy right so uh
12:00
friends right click deploy right so uh
12:00
friends right click deploy right so uh uh this is very much around
12:03
uh this is very much around
12:03
uh this is very much around um reducing the potential of human
12:06
um reducing the potential of human
12:06
um reducing the potential of human errors streamlining management task by
12:09
errors streamlining management task by
12:09
errors streamlining management task by just having everything codifying
12:11
just having everything codifying
12:11
just having everything codifying everything we do right so right you and
12:14
everything we do right so right you and
12:14
everything we do right so right you and I we don't deploy right we actually we
12:16
I we don't deploy right we actually we
12:16
I we don't deploy right we actually we shouldn't have access to production
12:18
shouldn't have access to production
12:18
shouldn't have access to production environments the first thing you do is
12:20
environments the first thing you do is
12:20
environments the first thing you do is set up a pipeline that that deploys an
12:22
set up a pipeline that that deploys an
12:22
set up a pipeline that that deploys an application the application itself is
12:24
application the application itself is
12:24
application the application itself is empty there's nothing there it it way
12:26
empty there's nothing there it it way
12:26
empty there's nothing there it it way you manage to deploy with automation an
12:28
you manage to deploy with automation an
12:28
you manage to deploy with automation an empty application that does nothing good
12:30
empty application that does nothing good
12:30
empty application that does nothing good for you now you put the application in
12:33
for you now you put the application in
12:33
for you now you put the application in there
12:34
there
12:34
there right yeah exactly and and I I I know
12:38
right yeah exactly and and I I I know
12:38
right yeah exactly and and I I I know I'm preaching to the choir in this show
12:39
I'm preaching to the choir in this show
12:39
I'm preaching to the choir in this show right but but it's just essential and
12:42
right but but it's just essential and
12:42
right but but it's just essential and that we keep that uh keep talking about
12:45
that we keep that uh keep talking about
12:45
that we keep that uh keep talking about the the manageability in in our
12:48
the the manageability in in our
12:48
the the manageability in in our architecture so I have a question about
12:50
architecture so I have a question about
12:50
architecture so I have a question about that have you felt that that was a big
12:54
that have you felt that that was a big
12:54
that have you felt that that was a big um hurdle of knowledge uh for the
12:57
um hurdle of knowledge uh for the
12:57
um hurdle of knowledge uh for the organization I you don't have to answer
12:59
organization I you don't have to answer
12:59
organization I you don't have to answer specifically but but I I have
13:01
specifically but but I I have
13:01
specifically but but I I have experiences that say that this is
13:02
experiences that say that this is
13:03
experiences that say that this is something that people are not always
13:05
something that people are not always
13:05
something that people are not always accustomed to out there because not
13:06
accustomed to out there because not
13:07
accustomed to out there because not every company has been doing that in in
13:09
every company has been doing that in in
13:09
every company has been doing that in in the in the past yeah this is this is
13:11
the in the past yeah this is this is
13:11
the in the past yeah this is this is hard yeah I mean we can take it out of
13:14
hard yeah I mean we can take it out of
13:14
hard yeah I mean we can take it out of the context of my work and just talk
13:16
the context of my work and just talk
13:16
the context of my work and just talk about in general and I would say yes uh
13:19
about in general and I would say yes uh
13:19
about in general and I would say yes uh it is uh it is a new way of working yeah
13:22
it is uh it is a new way of working yeah
13:22
it is uh it is a new way of working yeah for some people and if you're a software
13:26
for some people and if you're a software
13:26
for some people and if you're a software developer uh you might not necessarily
13:29
developer uh you might not necessarily
13:29
developer uh you might not necessarily uh know how to set up pipelines right
13:31
uh know how to set up pipelines right
13:31
uh know how to set up pipelines right it's a it's a skill it's a trade
13:33
it's a it's a skill it's a trade
13:33
it's a it's a skill it's a trade skill um and and that's why some teams
13:36
skill um and and that's why some teams
13:36
skill um and and that's why some teams they they invest in in those
13:38
they they invest in in those
13:38
they they invest in in those capabilities but but I I think we are we
13:42
capabilities but but I I think we are we
13:42
capabilities but but I I think we are we are getting into a much more um lean
13:45
are getting into a much more um lean
13:45
are getting into a much more um lean position and where I work at the lab
13:47
position and where I work at the lab
13:47
position and where I work at the lab group I mean we expect all product teams
13:50
group I mean we expect all product teams
13:50
group I mean we expect all product teams uh we are we are organized not like a
13:52
uh we are we are organized not like a
13:52
uh we are we are organized not like a functional it uh department but more
13:55
functional it uh department but more
13:55
functional it uh department but more like a a digital native company with
13:57
like a a digital native company with
13:57
like a a digital native company with product teams yeah working in their own
14:00
product teams yeah working in their own
14:00
product teams yeah working in their own bounded context and every team and we
14:03
bounded context and every team and we
14:03
bounded context and every team and we have more than 200 they are expected to
14:05
have more than 200 they are expected to
14:05
have more than 200 they are expected to be devops teams so they have to uh uh
14:09
be devops teams so they have to uh uh
14:09
be devops teams so they have to uh uh build and deploy their own code to
14:11
build and deploy their own code to
14:11
build and deploy their own code to production yeah and you can't do that at
14:13
production yeah and you can't do that at
14:14
production yeah and you can't do that at scale if you don't
14:16
scale if you don't
14:16
scale if you don't automate it doesn't work agreed agreed
14:19
automate it doesn't work agreed agreed
14:19
automate it doesn't work agreed agreed no that's not going to be possible um
14:22
no that's not going to be possible um
14:22
no that's not going to be possible um okay brilliant I like that I like that a
14:24
okay brilliant I like that I like that a
14:24
okay brilliant I like that I like that a lot what's the fourth are we up to four
14:26
lot what's the fourth are we up to four
14:26
lot what's the fourth are we up to four yes what's the fourth yeah the fourth
14:27
yes what's the fourth yeah the fourth
14:27
yes what's the fourth yeah the fourth one let's do the fourth one
14:29
one let's do the fourth one
14:29
one let's do the fourth one yes
14:31
yes
14:31
yes observable observable observable yeah so
14:34
observable observable observable yeah so
14:34
observable observable observable yeah so maybe that one requires also a bit of
14:37
maybe that one requires also a bit of
14:37
maybe that one requires also a bit of unpacking yeah so uh in a world of
14:40
unpacking yeah so uh in a world of
14:40
unpacking yeah so uh in a world of Highly
14:41
Highly
14:41
Highly automated
14:43
automated
14:43
automated um elastic application that are Loosely
14:46
um elastic application that are Loosely
14:46
um elastic application that are Loosely coupled right and you also talked about
14:48
coupled right and you also talked about
14:48
coupled right and you also talked about microservices as as a pattern you would
14:50
microservices as as a pattern you would
14:50
microservices as as a pattern you would have a lot of smaller
14:53
have a lot of smaller
14:53
have a lot of smaller applications uh I mean doing classic it
14:56
applications uh I mean doing classic it
14:56
applications uh I mean doing classic it monitoring is not going to do it right
14:59
monitoring is not going to do it right
14:59
monitoring is not going to do it right uh also as your application landscape
15:01
uh also as your application landscape
15:01
uh also as your application landscape Grows Right uh one team will not know
15:03
Grows Right uh one team will not know
15:03
Grows Right uh one team will not know what another one does but they are
15:05
what another one does but they are
15:05
what another one does but they are dependent right the apis the messaging
15:07
dependent right the apis the messaging
15:07
dependent right the apis the messaging is depending on each other so we simply
15:09
is depending on each other so we simply
15:10
is depending on each other so we simply need to invest in observability and this
15:12
need to invest in observability and this
15:12
need to invest in observability and this is about collecting and analyzing data
15:14
is about collecting and analyzing data
15:14
is about collecting and analyzing data to maintain system
15:16
to maintain system
15:16
to maintain system Health uh and we need both real time
15:19
Health uh and we need both real time
15:19
Health uh and we need both real time here but we also sort of need to look at
15:21
here but we also sort of need to look at
15:22
here but we also sort of need to look at this as um as a data set that we need to
15:25
this as um as a data set that we need to
15:25
this as um as a data set that we need to expose right because if I'm depending on
15:28
expose right because if I'm depending on
15:28
expose right because if I'm depending on on Magnus API in order for me to fulfill
15:31
on Magnus API in order for me to fulfill
15:31
on Magnus API in order for me to fulfill orders lot of people do yeah
15:34
orders lot of people do yeah
15:34
orders lot of people do yeah exactly um and and you cannot signal me
15:37
exactly um and and you cannot signal me
15:37
exactly um and and you cannot signal me if you're running degraded uh you
15:39
if you're running degraded uh you
15:39
if you're running degraded uh you wouldn't of course or if you're offline
15:41
wouldn't of course or if you're offline
15:41
wouldn't of course or if you're offline you wouldn't of course Magnus but but if
15:43
you wouldn't of course Magnus but but if
15:43
you wouldn't of course Magnus but but if we don't have that signaling in our
15:45
we don't have that signaling in our
15:45
we don't have that signaling in our Loosely coupled architecture yeah then I
15:48
Loosely coupled architecture yeah then I
15:48
Loosely coupled architecture yeah then I would start troubleshooting my own right
15:50
would start troubleshooting my own right
15:50
would start troubleshooting my own right uh because something is broken and and
15:53
uh because something is broken and and
15:53
uh because something is broken and and we when we have the three others we need
15:55
we when we have the three others we need
15:55
we when we have the three others we need observability as well here yeah some
15:58
observability as well here yeah some
15:58
observability as well here yeah some brilliant Solutions out there uh but it
16:01
brilliant Solutions out there uh but it
16:01
brilliant Solutions out there uh but it for many uh organization this is
16:04
for many uh organization this is
16:04
for many uh organization this is something that you you end up doing
16:06
something that you you end up doing
16:06
something that you you end up doing after the fact and we we need it as part
16:08
after the fact and we we need it as part
16:09
after the fact and we we need it as part of our our architecture we need to think
16:11
of our our architecture we need to think
16:11
of our our architecture we need to think big data in this as well and you need to
16:14
big data in this as well and you need to
16:14
big data in this as well and you need to be able to observe not only your
16:15
be able to observe not only your
16:15
be able to observe not only your application and how it the health of it
16:17
application and how it the health of it
16:17
application and how it the health of it but you are you are using and you will
16:19
but you are you are using and you will
16:19
but you are you are using and you will be using a lot of various um other
16:22
be using a lot of various um other
16:22
be using a lot of various um other services in the cloud for example in the
16:24
services in the cloud for example in the
16:24
services in the cloud for example in the in the past you might have an app server
16:27
in the past you might have an app server
16:27
in the past you might have an app server and and a database server with a piece
16:29
and and a database server with a piece
16:29
and and a database server with a piece of cable this long between them right
16:31
of cable this long between them right
16:31
of cable this long between them right but now that they are instead your
16:33
but now that they are instead your
16:33
but now that they are instead your application is hosted somewhere in an
16:35
application is hosted somewhere in an
16:35
application is hosted somewhere in an app server so similar but your database
16:37
app server so similar but your database
16:37
app server so similar but your database might be in a database service somewhere
16:40
might be in a database service somewhere
16:40
might be in a database service somewhere so you need to be able to observe that
16:42
so you need to be able to observe that
16:42
so you need to be able to observe that service as well to understand if your
16:44
service as well to understand if your
16:44
service as well to understand if your app is behaving strangely could it be
16:47
app is behaving strangely could it be
16:47
app is behaving strangely could it be that there's a hiccup in the service
16:49
that there's a hiccup in the service
16:49
that there's a hiccup in the service that you are in fact dependent on yes
16:51
that you are in fact dependent on yes
16:51
that you are in fact dependent on yes it's not only your code but it's it's
16:52
it's not only your code but it's it's
16:53
it's not only your code but it's it's those things as well yeah you need to
16:56
those things as well yeah you need to
16:56
those things as well yeah you need to think about analytics you need to think
16:58
think about analytics you need to think
16:58
think about analytics you need to think about money monor in real time
16:59
about money monor in real time
16:59
about money monor in real time monitoring you need to think about uh
17:02
monitoring you need to think about uh
17:02
monitoring you need to think about uh alerting so how do you activate your
17:04
alerting so how do you activate your
17:04
alerting so how do you activate your engineers when things go wrong because
17:07
engineers when things go wrong because
17:07
engineers when things go wrong because often we go into having business
17:09
often we go into having business
17:09
often we go into having business critical processes right and we need to
17:11
critical processes right and we need to
17:11
critical processes right and we need to be ready we need to do our fire drills
17:13
be ready we need to do our fire drills
17:13
be ready we need to do our fire drills so when things break I mean we have been
17:16
so when things break I mean we have been
17:16
so when things break I mean we have been in this industry for a while things
17:18
in this industry for a while things
17:18
in this industry for a while things break right things yeah could be your
17:21
break right things yeah could be your
17:21
break right things yeah could be your fault but it could be somebody else's
17:22
fault but it could be somebody else's
17:22
fault but it could be somebody else's fault you're dependent on something and
17:24
fault you're dependent on something and
17:24
fault you're dependent on something and that breaks right what happens to your
17:27
that breaks right what happens to your
17:27
that breaks right what happens to your application when that other thing breaks
17:29
application when that other thing breaks
17:29
application when that other thing breaks you know Magnus you know it's always DNS
17:32
you know Magnus you know it's always DNS
17:32
you know Magnus you know it's always DNS right yeah it's always DNS obviously
17:35
right yeah it's always DNS obviously
17:35
right yeah it's always DNS obviously come on you to identify which one it is
17:38
come on you to identify which one it is
17:38
come on you to identify which one it is right but it actually leads to the fifth
17:40
right but it actually leads to the fifth
17:40
right but it actually leads to the fifth characteristic which is
17:43
characteristic which is
17:43
characteristic which is resilience um so uh part of building
17:47
resilience um so uh part of building
17:47
resilience um so uh part of building Cloud native um is the ability to handle
17:52
Cloud native um is the ability to handle
17:52
Cloud native um is the ability to handle and
17:52
and
17:52
and recover uh failures gracefully and I
17:57
recover uh failures gracefully and I
17:57
recover uh failures gracefully and I find this as one of the coolest thing
17:59
find this as one of the coolest thing
17:59
find this as one of the coolest thing about working with h Cloud technology is
18:02
about working with h Cloud technology is
18:02
about working with h Cloud technology is that you actually have the components
18:04
that you actually have the components
18:04
that you actually have the components right we just need to think differently
18:06
right we just need to think differently
18:06
right we just need to think differently when we architect it um and often in a
18:09
when we architect it um and often in a
18:10
when we architect it um and often in a in a classic uh world of virtual
18:11
in a classic uh world of virtual
18:11
in a classic uh world of virtual machines we would we would then uh use
18:14
machines we would we would then uh use
18:14
machines we would we would then uh use hyper varus and do hardware for to so so
18:18
hyper varus and do hardware for to so so
18:18
hyper varus and do hardware for to so so kind of look at it bottom up because it
18:20
kind of look at it bottom up because it
18:20
kind of look at it bottom up because it will you Hardware is is a natural source
18:24
will you Hardware is is a natural source
18:24
will you Hardware is is a natural source of error right because you have mov
18:26
of error right because you have mov
18:26
of error right because you have mov Parts uh you have aging part and it
18:29
Parts uh you have aging part and it
18:29
Parts uh you have aging part and it breaks yeah but in the cloud world you
18:31
breaks yeah but in the cloud world you
18:31
breaks yeah but in the cloud world you don't have to worry about that layer uh
18:34
don't have to worry about that layer uh
18:34
don't have to worry about that layer uh but you still have failures right uh so
18:36
but you still have failures right uh so
18:36
but you still have failures right uh so this is about architecting for failures
18:39
this is about architecting for failures
18:39
this is about architecting for failures and there some great learnings in chaos
18:40
and there some great learnings in chaos
18:40
and there some great learnings in chaos engineering but it starts with good
18:43
engineering but it starts with good
18:43
engineering but it starts with good architecture so how do we reduce uh the
18:46
architecture so how do we reduce uh the
18:46
architecture so how do we reduce uh the impact of failures uh how do we even
18:49
impact of failures uh how do we even
18:49
impact of failures uh how do we even architect for higher
18:51
architect for higher
18:51
architect for higher availability um there's some really cool
18:54
availability um there's some really cool
18:54
availability um there's some really cool uh learnings here uh on on how we
18:57
uh learnings here uh on on how we
18:57
uh learnings here uh on on how we architect for uh for resilience
19:01
architect for uh for resilience
19:01
architect for uh for resilience yeah good yeah no that's that makes
19:04
yeah good yeah no that's that makes
19:04
yeah good yeah no that's that makes sense so we have all these things that
19:05
sense so we have all these things that
19:05
sense so we have all these things that we need to uh consider or to think about
19:09
we need to uh consider or to think about
19:09
we need to uh consider or to think about and they're not Tech dependent even at
19:12
and they're not Tech dependent even at
19:12
and they're not Tech dependent even at all these are these are strategies and
19:15
all these are these are strategies and
19:15
all these are these are strategies and patterns and and and ways of working
19:19
patterns and and and ways of working
19:19
patterns and and and ways of working even like devops is a way of working
19:21
even like devops is a way of working
19:21
even like devops is a way of working right um You can either deploy with
19:23
right um You can either deploy with
19:23
right um You can either deploy with automation or you can deploy manually
19:25
automation or you can deploy manually
19:25
automation or you can deploy manually right manually manually will suck you
19:28
right manually manually will suck you
19:28
right manually manually will suck you know it's going to take take time it's
19:29
know it's going to take take time it's
19:29
know it's going to take take time it's going to be error prone it's going to be
19:31
going to be error prone it's going to be
19:31
going to be error prone it's going to be horrible it might be easier just now
19:33
horrible it might be easier just now
19:33
horrible it might be easier just now right like right click deploy but I mean
19:37
right like right click deploy but I mean
19:37
right like right click deploy but I mean it's quickly eaten by you losing track
19:40
it's quickly eaten by you losing track
19:40
it's quickly eaten by you losing track of your changes ET absolutely absolutely
19:43
of your changes ET absolutely absolutely
19:43
of your changes ET absolutely absolutely so as you can see uh we have moved from
19:46
so as you can see uh we have moved from
19:46
so as you can see uh we have moved from well where things are hosted into how we
19:48
well where things are hosted into how we
19:48
well where things are hosted into how we host right using these five
19:50
host right using these five
19:50
host right using these five characteristics of cloud and I can I can
19:53
characteristics of cloud and I can I can
19:53
characteristics of cloud and I can I can only recommend it because it really
19:55
only recommend it because it really
19:55
only recommend it because it really gives the guidance to to our engineering
19:57
gives the guidance to to our engineering
19:57
gives the guidance to to our engineering colleagues that does the AR ecture right
19:59
colleagues that does the AR ecture right
19:59
colleagues that does the AR ecture right to to consider when you build build with
20:02
to to consider when you build build with
20:02
to to consider when you build build with these five and you will be all for a
20:04
these five and you will be all for a
20:04
these five and you will be all for a good start absolutely yeah it can be a
20:06
good start absolutely yeah it can be a
20:07
good start absolutely yeah it can be a journey to get to to proficiency on on
20:09
journey to get to to proficiency on on
20:09
journey to get to to proficiency on on all these aspects obviously because
20:12
all these aspects obviously because
20:12
all these aspects obviously because that's it's going to be hard right but
20:13
that's it's going to be hard right but
20:13
that's it's going to be hard right but it it will be
20:15
it it will be
20:15
it it will be rewarding it will it will all right
20:17
rewarding it will it will all right
20:17
rewarding it will it will all right brilliant that was uh absolutely
20:19
brilliant that was uh absolutely
20:19
brilliant that was uh absolutely excellent I am so happy that you came to
20:21
excellent I am so happy that you came to
20:21
excellent I am so happy that you came to the cloud show today to talk to us about
20:23
the cloud show today to talk to us about
20:23
the cloud show today to talk to us about how Lego group is going from being
20:26
how Lego group is going from being
20:26
how Lego group is going from being having a cloud first strategy to
20:28
having a cloud first strategy to
20:28
having a cloud first strategy to actually being Cloud native and what
20:31
actually being Cloud native and what
20:31
actually being Cloud native and what that actually
20:32
that actually
20:32
that actually means for having me Magnus absolute
20:34
means for having me Magnus absolute
20:34
means for having me Magnus absolute pleasure to be here it was my pleasure
20:36
pleasure to be here it was my pleasure
20:36
pleasure to be here it was my pleasure completely thank you thank you so much
20:38
completely thank you thank you so much
20:38
completely thank you thank you so much and uh audience thank you for watching
20:40
and uh audience thank you for watching
20:40
and uh audience thank you for watching the cloud show and uh see you next time
20:43
the cloud show and uh see you next time
20:43
the cloud show and uh see you next time [Music]