Welcome to a special BriefingsDirect panel discussion in conjunction with
The Open Group's upcoming conference on July 20, 2015 in Baltimore. Our panel of experts examines how cloud
governance and
enterprise architecture can prove essential in the Bimodal IT era, a period of increasingly fragmented IT.
Not
only are IT organizations dealing with so-called
shadow IT and myriad
proof-of-concept affairs, there is now a strong rationale for fostering
what
Gartner calls
Bimodal IT.
There's a strong case to be made for exploiting the strengths of
several different flavors of IT, except that -- at the same time --
businesses are asking IT in total to be faster, better, and cheaper.
The
topic before us then is how to allow for the benefits of Bimodal IT -- or
even
multimodal IT -- but without IT fragmentation leading to fractured
and even broken businesses.
Here to update us on the work of
The Open Group Cloud Governance initiatives and working groups and to further explore the ways that companies can better manage and thrive with
hybrid IT are our guests:
- Dr. Chris Harding, Director for Interoperability and Cloud Computing Forum Director at The Open Group.
- David Janson,
Executive IT Architect and Business Solutions Professional with the IBM Industry Solutions Team for Central and Eastern Europe and a leading
contributor to The Open Group Cloud Governance Project.
- Nadhan, HP Distinguished Technologist and Cloud Adviser and Co-Chairman of The Open Group Cloud Governance Project.
The discussion is moderated by me,
Dana Gardner, Principal Analyst at
Interarbor Solutions.
Here are some excerpts:
Gardner:
Before we get into an update on The Open Group Cloud Governance
initiatives, in many ways over the past decades IT has always been
somewhat fragmented. Very few companies have been able to keep all their
IT oars rowing in the same direction, if you will. But today things
seem to be changing so rapidly that some
degree of disparate IT methods are necessary. We might even think of old
IT and new IT, and this may even be desirable.
But
what are the trends that are driving this need for a multimodal IT?
What's accelerating the need for different types of IT, and how can we
think about retaining a common governance, and even a frameworks-driven
enterprise architecture umbrella, over these IT elements?
Nadhan:
Basically, the change that we're going through is really driven by the
business. Business today has much more rapid access to the services that
IT has traditionally provided. Business has a need to react to its own
customers in a much more agile manner than they were traditionally used
to.
We now have to react to demands where we're talking
days and weeks instead of months and years. Businesses today have a
choice. Business units are no longer dependent on the traditional IT to
avail themselves of the services provided. Instead, they can go out and
use the services that are available external to the enterprise.
To a great extent, the advent of
social media
has also resulted in direct customer feedback on the sentiment from the
external customer that businesses need to react to. That is actually
changing the timelines. It is requiring IT to be delivered at the pace
of business. And the very definition of IT is undergoing a change, where
we need to have the right paradigm, the right technology, and the right
solution for the right business function and therefore the right
application.
Since the choices have increased with the
new style of IT, the manner in which you pair them up, the solutions
with the problems, also has significantly changed. With more choices,
come more such pairs on which solution is right for which problem.
That's really what has caused the change that we're going through.
With more choices, come more such pairs on which solution is right for which problem.
A
change of this magnitude requires governance that goes across building
up on the traditional governance that was always in play, requiring
elements like cloud to have governance that is more specific to
solutions that are in the cloud across the whole lifecycle of cloud
solutions deployment.
Gardner: David, do you
agree that this seems to be a natural evolution, based on business
requirements, that we basically spin out different types of IT within
the same organization to address some of these issues around agility? Or
is this perhaps a bad thing, something that’s unnatural and should be
avoided?
Janson: In many ways, this follows a
repeating pattern we've seen with other kinds of transformations in
business and IT. Not to diminish the specifics about what we're looking
at today, but I think there are some repeating patterns here.
There
are new disruptive events that compete with the status quo. Those
things that have been optimized, proven, and settled into sort of a
consistent groove can compete with each other. Excitement about the new
value that can be produced by new approaches generates momentum, and so
far this actually sounds like a healthy state of vitality.
Good governance
However,
one of the challenges is that the excitement potentially can lead to
overlooking other important factors, and that’s where I think good
governance practices can help.
For
example, governance helps remind people about important durable
principles that should be guiding their decisions, important
considerations that we don’t want to forget or under-appreciate as we
roll through stages of change and transformation.
Attend The Open Group Baltimore 2015
July 20-23, 2015
Register Here
At
the same time, governance practices need to evolve so that it can adapt
to new things that fit into the governance framework. What are those
things and how do we govern those? So governance needs to evolve at the
same time.
There is a pattern here with some specific
things that are new today, but there is a repeating pattern as well,
something we can learn from.
Gardner: Chris
Harding, is there a built-in capability with cloud governance that
anticipates some of these issues around different styles or flavors or
even velocity of IT innovation that can then allow for that innovation
and experimentation, but then keep it all under the same umbrella with a
common management and visibility?
Harding:
There are a number of forces at play here, and there are three separate
trends that we've seen, or at least that I have observed, in discussions
with members within The Open Group that relate to this.
The first is one that Nadhan mentioned, the
possibility of outsourcing IT. I remember a member’s meeting a few years
ago, when one of our members who worked for a company that was starting
a cloud brokerage activity happened to mention that two major clients
were going to do away with their IT departments completely and just go
for
cloud brokerage.
You could see the jaws drop around the table, particularly with the
representatives who were from company corporate IT departments.
Of
course, cloud brokers haven’t taken over from corporate IT, but there
has been that trend toward things moving out of the enterprise to bring
in IT services from elsewhere.
That’s all very well to
do that, but from a governance perspective, you may have an easy life
if you outsource all of your IT to a broker somewhere, but if you fail
to comply with regulations, the broker won’t go to jail; you will go to
jail.
So you need to make sure that you retain control
at the governance level over what is happening from the point of view
of compliance. You probably also want to make sure that your
architecture principles are followed and retain governance control to
enable that to happen. That’s the first trend and the governance
implication of it.
In response to that, a second trend
that we see is that IT departments have reacted often by becoming quite
like brokers themselves -- providing services, maybe providing hybrid
cloud services or private cloud services within the enterprise, or maybe
sourcing cloud services from outside. So that’s a way that IT has moved
in the past and maybe still is moving.
Third trend
The third trend that we're seeing in some cases is that multi-discipline teams within
line of business
divisions, including both business people and technical people, address
the business problems. This is the way that some companies are
addressing the need to be on top of the technology in order to innovate
at a business level. That is an interesting and, I think, a very healthy
development.
So maybe, yes, we are seeing a bimodal
splitting in IT between the traditional IT and the more flexible and
agile IT, but maybe you could say that that second part belongs really
in the line of business departments -- rather than in the IT departments.
That's at least how I see it.
Nadhan: I'd like
to build on a point that David made earlier about repeating patterns. I
can relate to that very well within The Open Group, speaking about the
Cloud Governance Project. Truth be told, as we continue to evolve the
content in cloud governance, some of the seeding content actually came
from the
SOA Governance Project
that The Open Group worked on a few years back. So the point David made
about the repeating patterns resonates very well with that particular
case in mind.
I think there's a repeating pattern here of new approaches, new ways of doing things, coming into the picture.
Gardner:
So we've been through this before. When there is change and disruption,
sometimes it’s required for a new version of methodologies and best
practices to emerge, perhaps even associated with specific technologies.
Then, over time, we see that folded back in to IT in general, or maybe
it’s pushed back out into the business, as Chris alluded to.
My
question, though, is how we make sure that these don’t become
disruptive and negative influences over time. Maybe governance and
enterprise architecture principles can prevent that. So is there
something about the cloud governance, which I think really anticipates a
hybrid model, particularly a cloud hybrid model, that would be germane
and appropriate for a hybrid IT environment?
David Janson, is there a cloud governance benefit in managing hybrid IT?
Janson:
There most definitely is. I tend to think that hybrid IT is probably
where we're headed. I don’t think this is avoidable. My editorial
comment upon that is that’s an unavoidable direction we're going in.
Part of the reason I say that is I think there's a repeating pattern
here of new approaches, new ways of doing things, coming into the
picture.
 |
| Janson |
And then some balancing acts goes on, where
people look at more traditional ways versus the new approaches people
are talking about, and eventually they look at the strengths and
weaknesses of both.
There's going to be some
disruption, but that’s not necessarily bad. That’s how we drive change
and transformation. What we're really talking about is making sure the
amount of disruption is not so counterproductive that it actually moves
things backward instead of forward.
I don’t mind a
little bit of disruption. The governance processes that we're talking
about, good governance practices, have an overall life cycle that things
move through. If there is a way to apply governance, as you work
through that life cycle, at each point, you're looking at the particular
decision points and actions that are going to happen, and make sure
that those decisions and actions are well-informed.
We
sometimes say that governance helps us do the right things right. So
governance helps people know what the right things are, and then the
right way to do those things.
Bimodal IT
Also,
we can measure how well people are actually adapting to those “right
things” to do. What’s “right” can vary over time, because we have
disruptive change. Things like we are talking about with Bimodal IT is
one example.
Within a narrower time frame in the
process lifecycle, there are points that evolve across that time frame
that have particular decisions and actions. Governance makes sure that
people are well informed as they're rolling through that about important
things they shouldn’t forget. It’s very easy to forget key things and
optimize for only one factor, and governance helps people remember that.
Also, just check to see whether we're getting the
benefits that people expected out of it. Coming back around and looking
afterward to see if we accomplish what we thought we would or did we get
off in the wrong direction. So it’s a bit like a steering mechanism or a
feedback mechanism, in it that helps keep the car on the road, rather
than going off in the soft shoulder. Did we overlook something
important? Governance is key to making this all successful.
Gardner:
Let’s return to The Open Group’s upcoming conference on July 20 in
Baltimore and also learn a bit more about what the Cloud Governance
Project has been up to. I think that will help us better understand how
cloud governance relates to these hybrid IT issues that we've been
discussing.
Nadhan, you are the co-chairman of the Cloud Governance Project. Tell us about what to expect in Baltimore with the concepts of
Boundaryless Information Flow, and then also perhaps an update on what the Cloud Governance Project has been up to.
Attend The Open Group Baltimore 2015
July 20-23, 2015
Register Here
Nadhan: When the Cloud Governance Project started, the first
question we challenged ourselves with was, what is it and why do we need
it, especially given that SOA governance, architecture governance, IT
governance, enterprise governance, in general are all out there with
frameworks. We actually detailed out the landscape with different
standards and then identified the niche or the domain that cloud
governance addresses.
After that, we went through and
identified the top five principles that matter for cloud governance to
be done right. Some of the obvious ones being that cloud is a business
decision, and the governance exercise should keep in mind whether it is
the right business decision to go to the cloud rather than just jumping
on the bandwagon. Those are just some examples of the foundational
principles that drive how cloud governance must be established and
exercised.
Subsequent to that, we have a lifecycle for
cloud governance defined and then we have gone through the process of
detailing it out by identifying and decoupling the governance process
and the process that is actually governed.
So there is
this concept of process pairs that we have going, where we've
identified key processes, key process pairs, whether it be the planning,
the architecture, reusing cloud service, subscribing to it,
unsubscribing, retiring, and so on. These are some of the defining
milestones in the life cycle.
We've actually put
together a template for identifying and detailing these process pairs,
and the template has an outline of the process that is being governed,
the key phases that the governance goes through, the desirable business
outcomes that we would expect because of the cloud governance, as well
as the associated metrics and the key roles.
Real-life solution
The
Cloud Governance Framework is actually detailing each one. Where we are
right now is looking at a real-life solution. The hypothetical could be
an actual business scenario, but the idea is to help the reader digest
the concepts outlined in the context of a scenario where such governance
is exercised. That’s where we are on the Cloud Governance Project.
Let
me take the opportunity to invite everyone to be part of the project to
continue it by subscribing to the right mailing list for cloud
governance within
The Open Group.
Gardner: Just for the benefit of our readers and listeners
who might not be that familiar with The Open Group, perhaps you could
give us a very quick overview -- its mission, its
charter, what we could expect at the Baltimore conference, and why
people should get involved, either directly by attending, or following
it on social media or the other avenues that The Open Group provides on
its website?
Until an Open Group standard is published, there is no official Open
Group position on the topic, and members will present their views at
conferences.
Harding: The
Open Group is a vendor-neutral consortium whose vision is Boundaryless
Information Flow. That is to say the idea that information should be
available to people within an enterprise, or indeed within an ecosystem
of enterprises, as and when needed, not locked away into silos.
We
hold main conferences, quarterly conferences, four times a year and
also regional conferences in various parts of the world in between
those, and we discuss a variety of topics.
In fact,
the main topics for the conference that we will be holding in July in
Baltimore are enterprise architecture and risk and security.
Architecture and security are two of the key things for which The Open
Group is known, Enterprise architecture, particularly with its
TOGAF Framework, is perhaps what The Open Group is best known for.
We've
been active in a number of other areas, and risk and security is one.
We also have started a new vertical activity on healthcare, and there
will be a track on that at the Baltimore conference.
There will be tracks on other topics too, including four sessions on
Open Platform 3.0.
Open Platform 3.0 is The Open Group initiative to address how
enterprises can gain value from new technologies, including cloud
computing, social computing, mobile computing, big data analysis, and
the Internet of Things.
We'll have a number of
presentations related to that. These will include, in fact, a
perspective on cloud governance, although that will not necessarily
reflect what is happening in the Cloud Governance Project. Until an Open
Group standard is published, there is no official Open Group position
on the topic, and members will present their views at conferences. So
we're including a presentation on that.
Lifecycle governance
There
is also a presentation on another interesting governance topic, which
is on Information Lifecycle Governance. We have a panel session on the
business context for Open Platform 3.0 and a number of other
presentations on particular topics, for example, relating to the new
technologies that Open Platform 3.0 will help enterprises to use.
There's always a lot going on at Open Group conferences, and that’s a brief flavor of what will happen at this one.
Gardner: Thank you. And I'd just add that there is more available at The Open Group website,
opengroup.org.
Going
to one thing you mentioned about a standard and publishing that
standard, is
there a roadmap that we could look to in order to anticipate the next
steps or milestones in the Cloud Governance Project? When would such a
standard emerge and when might we expect it?
Nadhan:
As I said earlier, the next step is to identify the business scenario
and apply it. I'm expecting, with the right level of participation, that
it will take another quarter, after which it would go through the
internal review with The Open Group and the company reviews for the
publication of the standard. Assuming we have that in another quarter,
Chris, could you please weigh in on what it usually takes, on average,
for those reviews before it gets published.
I want to step back and think about what are the changes to project-related processes that new approaches require.
Harding:
You could add on another quarter. It shouldn’t actually take that long,
but we do have a thorough review process. All members of The Open Group
are invited to participate. The document is posted for comment for, I
would think, four weeks, after which we review the comments and decide
what actually needs to be taken.
Certainly, it could
take only two months to complete the overall publication of the standard
from the draft being completed, but it’s safer to say about a quarter.
Gardner:
So a real important working document could be available in the second
half of 2015. Let’s now go back to why a cloud governance document and
approach is important when we consider the implications of Bimodal or multimodal IT.
One of things that Gartner says is that
Bimodal IT projects require new project management styles. They didn’t
say project management
products. They didn’t say, downloads or services
from a cloud provider. We're talking about
styles.
So
it seems to me that, in order to prevent the good aspects of Bimodal IT
to be overridden by negative impacts of chaos and the lack of
coordination that we're talking about, not about a product or a
download, we're talking about something that a working group and a
standards approach like the Cloud Governance Project can accommodate.
David,
why is it that you can’t buy this in a box or download it as a product?
What is it that we need to look at in terms of governance across
Bimodal IT and why is that appropriate for a style? Maybe the IT people
need to think differently about accomplishing this through technology
alone?
First question
Janson:
When I think of anything like a tool or a piece of software, the first
question I tend to have is what is that helping me do, because the tool
itself generally is not the be-all and end-all of this. What process is
this going to help me carry out?
So, before I would
think about tools, I want to step back and think about what are the
changes to project-related processes that new approaches require. Then
secondly, think about how can tools help me speed up, automate, or make
those a little bit more reliable?
It’s an easy thing
to think about a tool that may have some process-related aspects
embedded in it as sort of some kind of a magic wand that's going to
automatically make everything work well, but it’s the processes that the
tool could enable that are really the important decision. Then, the
tools simply help to carry that out more effectively, more reliably, and
more consistently.
We've always seen an evolution
about the processes we use in developing solutions, as well as tools.
Technology requires tools to adapt. As to the processes we use, as they
get more agile, we want to be more incremental, and see rapid
turnarounds in how we're developing things. Tools need to evolve with
that.
Once you've settled on some decisions about evolving those processes,
then we'll start looking for tools that help you automate, accelerate,
and make consistent and more reliable what those processes are.
But
I'd really start out from a governance standpoint, thinking about
challenging the idea that if we're going to make a change, how do we
know that it's really an appropriate one and asking some questions about
how we differentiate this change from just reinventing the wheel. Is
this an innovation that really makes a difference and isn't just change
for the sake of change?
Governance helps people
challenge their thinking and make sure that it’s actually a worthwhile
step to take to make those adaptations in project-related processes.
Once
you've settled on some decisions about evolving those processes, then
we'll start looking for tools that help you automate, accelerate, and
make consistent and more reliable what those processes are.
I
tend to start with the process and think of the technology second,
rather than the other way around. Where governance can help to remind
people of principles we want to think about. Are you putting the cart
before the horse? It helps people challenge their thinking a little bit
to be sure they're really going in the right direction.
Gardner: Of course, a lot of what you just mentioned pertains to enterprise architecture generally as well.
Nadhan,
when we think about Bimodal or multimodal IT, this to me is going to be
very variable from company to company, given their legacy, given their
existing style, the rate of adoption of cloud or other
software as a service (SaaS),
agile, or
DevOps
types of methods. So this isn’t something that’s going to be a
cookie-cutter. It really needs to be looked at company by company and
timeline by timeline.
Is this a vehicle for
professional services, for management consulting more than IT and
product? What is n the relationship between cloud governance, Bimodal
IT, and professional services?
Delineating systems
Nadhan:
It’s a great question Dana. Let me characterize Bimodal IT slightly
differently, before answering the question. Another way to look at
Bimodal IT, where we are today, is delineating systems of record and
systems of engagement.
In traditional IT, typically,
we're looking at the systems of record, and systems of engagement with
the social media and so on are in the live interaction. Those define the
continuously evolving, growing-by-the-second systems of engagement,
which results in the need for big data, security, and definitely the
cloud and so on.
The coexistence of both of these
paradigms requires the right move to the cloud for the right reason. So
even though they are the systems of record, some, if not most, do need
to get transformed to the cloud, but that doesn’t mean all systems of
engagement eventually get transformed to the cloud.
There are good reasons why you may actually want to leave certain systems of engagement the way they are.
There
are good reasons why you may actually want to leave certain systems of
engagement the way they are. The art really is in combining the
historical data that the systems of record have with the continual
influx of data that we get through the live channels of social media,
and then, using the right level of predictive analytics to get
information.
I said a lot in there just to
characterize the Bimodal IT slightly differently, making the point that
what really is at play, Dana, is a new style of thinking. It's a new
style of addressing the problems that have been around for a while.
But
a new way to address the same problems, new solutions, a new way of
coming up with the solution models would address the business problems
at hand. That requires an external perspective. That requires service
providers, consulting professionals, who have worked with multiple
customers, perhaps other customers in the same industry, and other
industries with a healthy dose of innovation.
That's where this is a new opportunity for professional services to work with the
CxOs, the enterprise architects, the CIOs to exercise the right business decision with the rights level of governance.
Because
of the challenges with the coexistence of both systems of record and
systems of engagement and harvesting the right information to make the
right business decision, there is a significant opportunity for
consulting services to be provided to enterprises today.
Drilling down
Gardner:
Before we close off I wanted to just drill down on one thing, Nadhan,
that you brought up, which is that ability to measure and know and then
analyze and compare.
One of the things that we've seen
with IT developing over the past several years as well is that the big
data capabilities have been applied to all the information coming out of
IT systems so that we can develop a steady state and understand those
systems of record, how they are performing, and compare and contrast in
ways that we couldn’t have before.
So on our last topic
for today, David Janson, how important is it for that measuring
capability in a governance context, and for organizations that want to
pursue Bimodal IT, but keep it governed and keep it from spinning out of
control? What should they be thinking about putting in place, the
proper
big data and
analytics and measurement and visibility apparatus and capabilities?
Janson:
That’s a really good question. One aspect of this is that, when I talk
with people about the ideas around governance, it's not unusual that the
first idea that people have about what governance is is about the
compliance or the policing aspect that governance can play. That sounds
like that’s interference, sand in the gears, but it really should be the
other way around.
Good governance has communicated that well enough, so that
people should actually move faster rather than slower. In other words,
there should be no surprises.
A governance
framework should actually make it very clear how people should be doing
things, what’s expected as the result at the end, and how things are
checked and measured across time at early stages and later stages, so
that people are very clear about how things are carried out and what
they are expected to do. So, if someone does use a governance-compliance
process to see if things are working right, there is no surprise, there
is no slowdown. They actually know how to quickly move through that.
Good
governance has communicated that well enough, so that people should
actually move faster rather than slower. In other words, there should be
no surprises.
Measuring things is very important,
because if you haven’t established the objectives that you're after and
some metrics to help you determine whether you're meeting those, then
it’s kind of an empty suit, so to speak, with governance. You express
some ideas that you want to achieve, but you have no way of knowing or
answering the question of how we know if this is doing what we want to
do. Metrics are very important around this.
We capture
metrics within processes. Then, for the end result, is it actually
producing the effects people want? That’s pretty important.
One
of the things that we have built into the Cloud Governance Framework is
some idea about what are the outcomes and the metrics that each of
these process pairs should have in mind. It helps to answer the
question, how do we know? How do we know if something is doing what we
expect? That’s very, very essential.
Gardner:
I am afraid we'll have to leave it there. We've been examining the role
of cloud governance and enterprise architecture and how they work
together in the era of increasingly fragmented IT. And we've seen how
The Open Group Cloud Governance Initiatives and Working Groups can help
allow for the benefits of Bimodal IT, but without necessarily IT
fragmentation leading to a fractured or broken business process around
technology and innovation.
Attend The Open Group Baltimore 2015
July 20-23, 2015
Register Here
This
special BriefingsDirect thought leadership panel discussion comes to
you in conjunction with The Open Group’s upcoming conference on July 20,
2015 in Baltimore. And it’s not too late to register on
The Open Group’s website or to follow the proceedings online and via social media
such as
Twitter and
LinkedIn.
Listen to the podcast. Find it on iTunes. Get the mobile app for iOS or Android. Read a full transcript or download a copy. Sponsor: The Open Group.
You may also be interested in:
