Transcript
This transcript was autogenerated. To make changes, submit a PR.
Hi everyone, thanks for joining us. Today I'm going to talk about building secure
architecture in the cloud. It's like the talk of the town wherein people
who have never thought of moving to cloud are talking about moving to
cloud. Now this is a situation happening because of the pandemic that we
are in, so why not talk about it? That how can we secure
the architecture in the cloud when it's the need of the hour? And it's not
just that the bigger organizations have to make sure that they
are secure, it goes for the startups as well. So today
I'm going to give you a brief architecture or the
knowledge that I have around building secure architecture in the cloud.
So while we deep dive into it, let me talk a bit about myself.
IBM, Vandana Verma Segur. My day job is with one of the multinational companies
where I am working as a product security architect. Apart from that,
I root role owner work wherein I support OwAs,
which is open web application security project. I'm one of the global board
of director for OwAs and I also work a lot
bringing diversity and information security, which doesn't
only need to be come, it can be anyone who wants to be part of
cybersecurity. It can be from different domains, it can be the person who's going
to be flipping domain from any other to information security.
So that's what I do during my free time.
I love thinking about topics that I know or I love researching
about new things in cybersecurity. Today's agenda is
more around cloud strategy. Why do we need to have an architecture
review for cloud when we are thinking about moving to cloud, or even if
we have already moved to cloud, what are the things that we need to consider?
At the same time, what are the scenarios that are stopping us from moving
to the cloud or when we have already tough, okay,
now this is the way to go. What are the considerations that we should have?
And then we will sum up all of that. So when I talk about
public cloud and the data that is moving to cloud,
that's a huge number and has per one of the research which
is latest, that 90% of the data and analytics
would be there in public cloud by 2022. And at
the same time when we are in pandemic, there are over 98%
people, 98% organizations which have moved to cloud.
That's one of the status. And some people were
planning to move to the cloud, but this situation has made us
so that we have taken the steps to move to the cloud. At the
same time, when I talk about the services that are part of the cloud,
solutions or public cloud. The services market is expected
to reach its new number, which is 354 billion
dollars, which is huge number. And 83%
of the enterprises workloads will be in cloud by
2020. That was the saying. But now when
we are in pandemic, that situation has reached another peak point.
Apart from that, 66% of the enterprises
already have a central cloud team or a cloud
center for excellence. They've started to things about it.
They want to have something that can maintain. They wanted to have a team which
can maintain the cloud. Now when we are talking about this numbers wherein
the organizations are spending so much on the cloud, talking about moving
to cloud, or have moved to the cloud and reaching its own big number
organisations are moving to the cloud,
sometimes the situation says, okay, you have a plan,
you're going to go to the cloud by this date. But now when we are
in a kind of a situation wherein we are dealing with a pandemic, organizations have
already fast forwarded their move to the cloud and more than 1.3
trillion it spending will be affected by this shift.
And always when we talk about these numbers, cloud adoption,
spending on the cloud, it has positive as well as negative impact.
There are organizations who have a view that, okay, this is what they want
in the cloud, but at the same time, there are organisations who have
a plan to move to cloud, but they don't know whether cloud is
the best solution or which cloud is the best solution for them. So these
are some numbers which are going to spill the beans and change the numbers
in the coming future. If I talk about why organizations
have concerns in moving to the cloud, or when a small startup
thinks that they're going to move to the cloud, but they are facing some
challenges. If I highlight some, it's because of the unknown territory
that everything is moving out of the organizations. And since
ages, all the data were always in the
data centers. But now when we are moving, the scenario
or the paradigm is shifting, cybersecurities are becoming day to day
struggle for businesses. Recent trends in cybersecurity statistics reveal
a huge increase in the cybersecurity space, like the organisations
getting hacked, data breaches. And not
just that. Now when we are living in the world of IoT,
Internet, connected devices, mobiles, there's another shift which is happening.
There's so much of concern around the data breach.
Additionally, recent security research suggests that most companies
have unprotected data poor cybersecurity practices in place.
There are teams who are working towards securing in organizations
where there's so much data still lying out there and there
are still vulnerable services on which
if we talk about touching, that takes a huge amount of time to
patch a vulnerability. If it exists, especially if it's an internal vulnerability,
people take time to resolve that. Another shift
that I've heard that people are moving their legacy applications to
containers or to cloud. But think about legacy applications already
have an overhead. Why do we need to move them to docker or
containers if we don't have a proper plan? They will still be an overhead if
we don't properly update it. Lifting and shifting never, always helps.
And when it comes to the legacy applications, that's what have been
the strategy for a lot of organizations. So we need to address
that concern as well, wherein if there is an application which is legacy application,
and we are trying to go ahead and move that to the cloud, we need
to make some changes so that it can be cloud ready. It shouldn't be just,
we are going to pick it up and post it there. It should not be
the copy and paste. There will be lot of overhead.
And not just that, there will be lot of budgeting constraints
that will come into picture, because to manage that
in the cloud, we would need resources and whatnot, and whether the application
will get the required kind of support would be there or not.
Who knows about that. Another concern is the hybrid cloud kind
of an environment or multicloud kind of an environment.
Organizations think that do we need a multicloud strategy,
hybrid cloud strategy, what kind of things that we are looking for?
So that's, again another kind of a scenario which needs to
be addressed. Because if we don't understand that vendor
lock in could be there, if one cloud doesn't suits us,
we can move to another cloud. So that's a very big area to address.
At the same time, regulatory concerns. There are so many regulations
that we have to follow, especially if in us, we talk about CCPA,
if it's California, if it's a european, then we have to talk about
GDPR. And then there are so many other compliance certifications
that are there, or requirements which are there, which needs to be addressed.
There are standards that needs to be adhered to. There's a governance mechanism that
we need. How are we going to attain that? That's a really
big area to be addressed. Now let's look at if we don't
have a right kind of architecture in the cloud, what could go wrong?
Managing data on the cloud is a function, actually a task
that is co owned and co managed by the cloud. Customer,
which could be us or could be an organisations and the cloud
service provider which is CSP. So it's a shared responsibility and
this adds up to efficiency and cost optimization at the same
time. But there are still few gray areas that we need to address and look
after. So as the saying goes, the chain is
only as strong as its weakest link, while a regular pen test
activity offers a certain level of assurance that yes,
we are safeguarded from the threat actors and our infrastructure is secure.
However, it may not go into the details of overall architecture
design and if we don't do the right architecture design, we could be in
trouble. So a secure review of architecture is very,
very important. It look at what's there in the internal network
and what are the nuts and bolts of a network. We also
need to have the information around our architecture.
We need to do some exercises to get to know our own network.
If we do not know, which is like security misconfiguration
or we have left a bucket open for anyone and everyone to see, the database
credentials are there in the code which is publicly available even when
you change it. The thing that is there on the code on public
that will be cached forever. And the code of public or
cloud Secure architecture review is to identify and highlight those
cloud security weaknesses and strengths and provide the
right guidance towards making a mature security
architect structure. Now with an increasing number of organisations
which are moving towards agile and DevOps methodologies and running
their infrastructure in the cloud, and they are trying to attain
the whole product itself, their organisations is based
out of the cloud. So it's really, really important to secure that
the attack surface has actually widened for them, and not
just for them, but overall it is widened and it's very,
very difficult to sometimes understand that what's our own
play area or what's our own environment now there has
been moving parts in that also wherein we are deploying changes into
the production itself directly. I have seen organisations deploying changes
to production, who does that? It's like the most unsafe thing to do,
that you're just putting an x on your own feet. So identifying the
attack surface and evaluating our own environment is the
first and foremost thing. And if we don't understand and
address these challenges, we're going to be in big, big trouble. So do
we need an architecture review? Think about it. Obviously we do
need it, because we have seen that there have been increase in the
cybercrimes and breaches and every other day we see that were
is an organization getting attacked or hacked or breached.
Now when that happens, as per one of the study,
when an organization gets breached. That doesn't happen for one,
like, that doesn't happen once. It happens a couple of times in
a short span. Now, the organization know that we have been
compromised, and still the same continues to happen,
because it's not that the organization don't safeguard them,
it's because sometimes we don't understand our own attack surface.
So it's very important to understand what's our protect surface,
which is like our own data network
assets or applications, or even our users. So it's
very important to understand these aspects. So if they don't
understand, what would be the impact. So the risk
of breaches can be huge. It can be that
organization could have penalties, which is due to regulatories,
like they've seen so many penalties due to GDPR or
customer data loss. This is like losing the customer
data and then losing the customer. So we can have a big problem,
and then the reputation, which is one of the most important thing
in brand value. Another thing would be once we lose one customer,
we tend to lose multiple at the same time. And when we talk about the
statistics that go a long way. So security breaches have
increased in huge number, I would say by 11% since 2018 and
67% since 2014. There's a huge number
or huge increase. Hackers attack every 39 seconds
on an average. Wow, that's gross. So the average time
to identify a breach in 2019 was
206 days. God, if an organization
figure out that much of time, that, yes, they have been compromised.
It's a big problem. So all these data have collected from the Ponymon
research done by IBM in partnership with the IBM.
So you can look at the report done in 2020 that has the
latest data about what could be the impact of
all these breaches that are happening now. When we talk about migration
to the cloud, that's actually a fear of unknown.
And cloud brings the Internet benefits and it brings the breaches also
at the same time. So building on top of the native security
of the cloud, which is there, will help us big time.
Sometimes we don't understand that, yes, this is there in the cloud and we can
take help from. I have seen it. So it's very,
very important to understand what our cloud security provider, which is providing.
So there are still so many questions that
are unanswered, or there are gray areas which are lingering around and which
lead to these exposures and breaches that happen.
Now, what are these? If I talk about these
questions, how my data look in terms of format,
state flow, and more when we talk about on premise and
cloud. But what do we need to understand? Like, how are we going to
safeguard my data when we talk about moving from on
premise to the cloud, what level of access do the
cloud security provider employees get to the data when I host
on the cloud, like whether they're going to access my data, and if I host
it on the cloud, that means they have access to anything and everything. So are
they segregating it or not? At the same time, if I talk
about cloud security providers native security features, which are great,
I've seen with so many vendors, there are great features, but how do
I know how much of them have been exercised or thought through,
or whether they are even effective? And how do I get that there
is an assurance? Yes, there are regulatory requirements
my cloud provider have met and there are auditors who have
come and measured that. So it's very important for me to understand. And how do
I get notified if there is a breach with that cloud provider,
or even if were is a breach attempt also?
So that's really, really a concerning area. And if I
talk about the research by Gartner. And Gartner predicts that
by 2023, the lead cloud service provider will have a distributive
ATM environment which have the presence to serve
many services and apprehensions to not take a backseat.
So all these things will be considered. Now,
if I talk about how should we do the architecture review?
Now, I have talked a lot about breaches. I've talked about
this could be the impact, but did I talk about architecture,
which is like, talk is all about. So now we're going to talk about architecture.
What are the lessons that we should be learning from that?
So first, and the foremost thing is that it's all about aligning the
fact that there are risks, understanding that there are risks when
we are moving to the cloud, the first thing that we have to understand that
our network is hostile, completely hostile, before even moving
to the cloud. So we start that, yes, now we have to secure our
environment, and that's where the first step lies. And when we
are there, we need to assess the risk and we have to have a proper
plan that how are we going to secure our own network. At the
same time, there are responsibility models that we are going to
be talking about. We need to understand them. That, yes, it's a shared
responsibility of a client and cloud service provider.
So it's a shared responsibility. I cannot say that I have hosted my data on
the cloud and I am all secure. No cloud security provider has
given you a platform, he has given you infrastructure, or he has given you the
software, or they've given you that, but your data is on your
own responsibility. I can't say that. I have a house
and I have my tenants, and my tenants are going to take care of my
full house. No, that's not going to be there. You have to manage your own
house. Tenants will be there. If something goes wrong, they're going to call you,
they're not the offers. So it's like that when we are moving to
cloud, it's like we are moving to a house which is we have rented.
So we need to understand the whole clauses which
are there in the agreement. It's very, very important to get the lawyers
and the people who understand the language, or the auditors who understand the
language. Yes. This is what this cloud provider is abiding by.
And these are the things that needed to be
considered while moving to the cloud. Now let me show you an architecture.
This is an IBM cloud architecture. And if I talk about AWS
or Azure, most of the clouds almost look similar,
wherein we have to address the concerns which are there as part of governance,
risk and compliance. Then we have to understand the different areas which
are there which talk about security in
the cloud, like identity and access management, application security, data security,
infrastructure security, or anything else. Like there are multiple layers.
People say that there are seven layers that we need to understand.
There are seven parts that we need to understand, and it's an integrated thing
wherein we address the physical security, the whole platform security,
even the services that we are getting in the cloud. So based
on the business needs, companies adopt for different models which
are part of the cloud, like whether they want to go for public
cloud, private cloud, or even kind of hybrid cloud. And now
a thing called multicloud. So organizations have started to talk
about it. And when we are selecting the appropriate cloud, okay, this is what I
need. We need to understand the deployment model, what kind of deployment model we will
be going by, and the type of application that we have,
the data sensitivity we have. We have to very much understand that and
the kind of business process importance, who will be
our target users. So it's very important. And if we
talk about single deployment or development and management exercises
across the platform, if I talk about public,
private and hybrid cloud, it's important to understand the basics
that what do we need? If I am okay to take the route
for public cloud, or if I'm okay to take the route of private or a
public cloud, or do I really want to be a
part of multicloud strategy. So it's all depend on organization.
And the most important thing that we need to understand is that we need to
have a proper plan to address the security components that are part
of our security architecture. If I talk about network security,
I need to understand that what will be my
boundary protections or if I'll have isolation or not,
what will be the network access control would be there, or boundary protection
would be there for my web application or not, whether I will also have
a firewall which is there for my internal resources. I have seen cases
where we are talking about web app firewall or WAF for
our external resources, but we tend to forget our
internal resources. If there is an internal attack happens,
for God's sake, nothing should happen. If that happens, how are
we going to detect that, especially on our internal applications? We have so
many internal applications in the organizations, could be HR,
could be internal employee apps, or even for
learning portal. So there are so many applications that we have,
or there are some mission critical applications that we have. So how
are we going to handle that situation? How are we going to address the concerns
which are there with the identity and access management? Especially now
when we are not in the four walls of an organization,
it becomes even more important to address these access
or authentication or authorization related
flaws. How about having multifactor authentication, and not
just for the people who are privileged users, but for everyone and
adopting the principle of least privilege, how are we
going to, can we have different authenticator management systems?
Like I'm not comfortable in getting OTP on my phone. I have
seen a lot of recent cases wherein it was
cloned and then attackers gained lot of money
out of it. So those are SIM card issues. But then again those
are concerns. And when we talk about cloud, cloud is like,
if I get your credentials, especially root credentials, I can own
your network. And the fun fact is that for
one of the project I was working, it's not a recent incident,
but sometime old incident wherein the
developers were using the root credentials and I was like, wow,
if these credentials go anywhere, what could go wrong? The whole
applications can go down, or somebody might have access to anything and
everything at the same time. We need to address the concerns which are there
for our own application and workload security like vulnerability
scanning, understanding the virtual services which are there,
or the ingress points that we have. And now when
we are talking about Kubernetes services or services,
we need to understand that, how are we going to provide the access to that?
And are we going to have role based access what kind of access
are we going to have? So it's very, very important to have. We also would
need to address the data security concerns, whether the data would
be encrypted at rest, motion or in
transit. Are we going to have TLS, endpoint security,
confidentiality and integrity, or are we going
to have encryption key management? So these are the things that
if we don't understand on, we're going to be in trouble.
And last but not the least, if we don't log it,
we can never safeguard ourselves. So how are we going to manage our
audit events, relevant audit logs, where are we going
to store it, for how long we are going to store it, how are we
going to protect our audit information? At the same time,
what infrastructure we would be monitoring, where the
audit records would be there for retention. So it's very, very important to
understand that. And when we talk about architecture services.
So this is the architecture which I picked up from one of the social media
sites. I don't know who posted it, but yes, I found it really relevant that
these are the things that we need to address, which starts with the
whole strategy, planning and then having the right governance,
doing the gap analysis or education, which is important,
the training for the teams are very important. Doing the right kind of risk assessment
is very, very important. So we need to have an architecture review methodology
which starts with data flow diagram. And even
before that, when I talk about the new normal that we are in,
businesses have to have a quicker and agile transition to the cloud,
which is with cloud enabled operations. This means thriving
a lesser known environment with unprecedented threat,
which I don't know what we are going to be doing with those threats in
the cloud. So these unprecedented times or
unknown times have taught us so many things about
our own environment. So this time has tough us a lot.
Were should we start off? We should first start off with the data flow diagram.
We need to understand our own cloud architecture infrastructure and
this will be done by referring to target organization, where we
need to have a high level cloud architecture. So data flow diagrams
actually help in that. Then we have to have a proper threat modeling done
of our cloud infrastructure and there has to be proper resources
that should be created for that. We also have to have a risk based plan
that, okay, this is how we are going to move to the cloud. These are
the things that we need to safeguard. This is where we are going to be
landing. And we need
to have adequate visibility on our environment and
understand our own environment with a gap analysis.
Where is the gap? How should we address that so let's
start with the threat modeling. Threat modeling is crucial, as it
sounds, or it help teams to proactively understand and
develop a strategy for identifying the potential threats
early. And for cloud, it's very, very important to do
the threat modeling of our own environment. There are multiple models which are
there like stride, pasta, trike, wasp. There are so
many. I have been working on stride a lot lately, and I
really like the kind of model stride is. And when we talk about
threat modeling, threat modeling can be done as a code. And threat modeling,
we need to understand that this is not just code as well. So the tool
is less important than the data recorded. So we
have to have data recorded in that and using a tool already.
Okay, keep doing so if you're doing it. Whiteboarding is
my favorite. So I always say that for threat modeling,
whiteboarding is like, amazing way to go. How are we
going to have the threat models that we have created for longer run?
Now, there's another thing that we have to understand. When we are doing the
threat modeling, we need to create data flow diagrams. Now, when we create
the data flow diagrams, we need to define our trust boundary.
And when we are creating those trust boundaries,
we need to understand the whole architecture, especially this new normal
demands businesses to have a quicker and agile transition.
So how are we going to connect the systems with the data flow and where
the data would be passing through? So it's very, very important to have
the whole map. So the advantages of utilizing these processes
or application flow diagrams, wherein we can create the threat models,
which are developer friendly, and wherein we can showcase, okay, these are the applications
we have. This is the whole architecture that we have.
And it can also help us in creating the process map, which shows how individuals
move through an application. Security professionals and developers can then view
that these applications, if I don't act upon time, how the
attacker might be envisioning them, which is more efficiently
prioritizing the potential threat. And if we have
a standardization of these threat modeling processes, it can help us in
consistent, actionable output, which can help us a long
way. And companies need to govern how data or
how connections to data sets are established, and we need to monitor
this data transfer. So data flow diagrams are the best way
to go. And employees, organizations need to understand
that the risk of data exposure is huge. And if we don't have the
right kind of training around these data flow diagrams, or in general about
cloud security, it can lead us to a wrong way. Now,
we've talked about threat modeling a bit, data flow diagrams a bit.
Now, let's talk about shared responsibility model that we were talking
about in the beginning. So shared responsibility model, talk about that.
What would be the responsibility of the cloud provider
and what would be the responsibility of a client or us,
let's say, to be precise. So identifying the security risk
management responsibility is very, very important. Clearly chartered
security responsibility model is the way to go.
So the Covid-19 pandemic changed the business practices
worldwide and the traditional workspace transformed
or changed for many, the new norm, like people are
saying, new normal, new norm, which is now a new home office.
We are all homebound, we are doing the virtual, we are in
a virtual workspace, all servers are on cloud and we are
meeting over different platforms like Slack hangout teams and
whatnot. So the cloud is busier than ever. Making cloud
security is more important than ever. So in public cloud,
there is a shared responsibility model wherein how
cloud security provider and we are going to manage it and security for things
like data classification, network controls,
physical security and whatnot. So if you can see on the screen,
we have talked a bit about data classification and accountability and
endpoint protection. We have talked about application level controls, network controls,
host infrastructure, and not to forget
physical security. If we don't understand and address
that, that can be a big concern. So based on whether we are running IIs
infrastructure as a service or pass implementation, which is platform as a service
implementation, we might want to certain the additional security
responsibilities that we have. So whether it's in the
data center, using a service, server based infrastructure as
a server instance or serverless system, or even a
past system, we are always responsible for securing our own
data. Talk about information and data. If we don't retain
our control over the information and data which
we maintain, we would never be able to understand what is
the visibility of our data. We will have zero visibility into
our data and all the data access in our control would
never be there. And again, another thing is identity and
access management. So how there are different facets to
it, different terms to it. So if we don't have the right IBM set,
which is identity and access management set, right kind of group set,
we don't have a single sign on mechanism or multi factor authentication, we're going to
be in trouble, big trouble. Apart from that, we need to understand the
application logic and code. So regardless of how we choose
to spin up the cloud resources, our proprietary applications are ours
to secure and the control throughout the application
lifecycle. If we have moved an application to the cloud, we cannot say that
now the cloud provider is going to secure our application? No, there can
be abuses that can happen. So this includes securing our own code
repositories from the malicious or misuse and
talking about application build testing throughout the
lifecycle, ensuring that secure production access is there and maintaining
security on any connected systems. Talk about virtual
machine accesses which are there in the cloud. So it's very important to
have a role based access control or started using the
principles of least privilege. So when we spin up
the cloud environment, we control the operation environment. And how
do we maintain or control those environments varies based on the kind
of instance that we are going for, whether we are talking about maintaining
the operating system. So we have to harden it. If we are maintaining the
applications, we have to make sure that when we are doing the devsecops,
we are doing it with the right things. So for the serverless
resources, the cloud providers control planes
give us the access to the setup of the configuration. So it's very, very important
to understand what's our control plane and the data
planes that will surround us and how the control plane will be
managing the configuration. And if a cloud provider is giving us the right to
manage that, there's nothing beautiful than that. And in case of server based
instance, we need to understand that. How are we going to manage the identity
and directory infrastructure? How are we going to manage our own applications?
How are we going to manage the network controls that are there? How are we
going to manage the operating system? I can go on and on. When we talk
about shared responsibility model, there's a huge thing to understand
and not to miss. We need to have a right kind of risk based
approach and if we don't have a plan, we can never attain
a right kind of shared responsibility model. It will just lie there and nothing
is going to happen. So what are the various data sets plan
to migrate to the cloud? Understand that. What are the regulatory
requirements from a trust border perspective? Like there
are data centers which are there. If I'm moving the data from one
data center to another, how are we going to manage those cross
border perspectives? And do we have policies and baseline
defined to be enforced on these workloads? How does the
data lifecycle look like? How are we going to manage our
own data? What are the encryption
measures which are there and how do we translate it to
the cloud? So the outcome would bring in a
clear and documented cloud security roadmap. If we
understand these things, if we have a plan, we understand shared responsibility
model. We address the identity and access management concerns and the
concerns which are part of the different layers that we have.
Last but not the least, gaining the right kind of visibility.
And if we have the right kind of visibility, we can have the right set
of service level agreements defined. What is actually
the need of the r for the business. So a due diligence. Diligence is
very, very important. And not to forget, we need to have
a kind of right logging and monitoring. We have to have an incident
reporting, a proper response plan
that has to be there. What will be the business continuity management if
one data center goes down, what are the legal implications that would be
there or if were are contract that I need to change
if I want to move from one vendor to another, what will be the legal
implications or if I want to move services from one cloud provider to the other
provider? And if there is a vendor lock in, how are
we going to move past that? So another important
thing is that uptime a right to audit, can we audit the
cloud provider? I have seen that cloud providers are not, a lot of cloud
providers are not let clients do the risks assessment. Instead they
give the reports which are there. So how are we going to understand that
these things are important for us, things which are not
to miss, which are latest things and people don't recognize
it and don't address it. So let's start with CASP. So Casp is
cloud access security broker which is required as a
central shared security thing which can be there
at cloud edge which can help in cloud related traffic monitoring
and preventing controls. Another thing would be data user
behavior and activity monitoring within and across the authorized
and unauthorized SaaS cloud providers. How about using
the shadow it and how about using those shadow cloud
and protect that? How are we going to have protection against the malwares?
At the same time, understanding the cloud infrastructure, configuration management,
container security, traffic management, threat management. So CASp is
the anchored multi cloud safety net is very, very important
for us to use. Now how about were going to address the challenges
which are there across the cloud security providers? We have to address the concerns for
identity, authorization and authentication mechanisms.
So we need to have a pragmatic approach towards implementing
all across the cloud providers. Another thing is that application security
basics like implementation, configuration and audit
of security design and configuration is very, very important,
especially if we have within SaaS has
pass and now functional as a service as well. So what
will be the management from the cloud security provider, end IBM configuration
or network configuration? What will they be managing?
What are the services that they are providing? So it's very very important to
address these concerns. When we talked about cloud security architecture. Now,
to sum it up, we need to look at the broader picture that we have.
We need to protect the customers, partners and our own business. We work with
a lot of partners. We need to think out of the box. We need to
have a different kind of understanding about our
own network. I have seen organizations who are moving to cloud, they don't
understand their environment well. They don't know what they have in their
environment currently, which is a big, big gap.
And if we address that, we address the skill
gap that is there. Another important aspect is that
when think about moving to cloud, we need to
do the design thinking. Like, okay, this is how my design is
going to look. We need to involve all the stakeholders which will be part of
this movement to cloud. There must be multiple teams who would
be moving to cloud. So changed is the only thing that is constant.
And when we talk about cybersecurity, it's very important to address these concerns.
Now, one more important thing that I want to tell
before we go that building a security in the cloud is not
an easy task, but not a difficult task. What we can do is
we can start with building the security architecture with zero trust.
Thank you so much. If you have any questions, any concerns,
any discussions you want to have, please do feel free
to ask me any questions. I am happy to have a conversation
on my twitter handle, which is infosec one Dana, or you can reach out to
me on LinkedIn. I'm there, Verma, and I'll be
more than happy to have a discussion around this topic. Thank you so much.