Winning Manufacturing Strategies

« July 2012 | Main | October 2012 »

September 26, 2012

The cloud applications recipe: the right proportions of public, private, hybrid (and a master chef)

So cloud computing is the newest cuisine in your enterprise IT menu. Chances are that you're sourcing the ingredients from many vendors and trying to use these ingredients in a way that creates the perfect cloud smorgasbord (buffet-style meal with multiple dishes). One that delights your business users and customers, and has them coming back for more.

You may have different sections in your cloud menu - applications, infrastructure, and platforms. You may have many considerations ― is it safe to consume (security)? Does it go well with your existing menu (does it integrate with existing assets)? Is it ready to be served (provisioning)? Can you run the same menu for a long time (sustenance)?

A word of caution though - too many cooks may spoil your cloud broth. Two imperatives, while you craft your cloud menu, are:

  1. Your applications menu is your piece-de-resistance: How you serve up your applications for use may end up making or breaking your cloud strategy. It's really important to ask yourselves: "What are we doing with applications?"
  2. You need a master chef to ensure cloud success: You may source your cloud ingredients from multiple technology providers, but so does everyone else. What will really set your cloud menu apart? The answer lies in having one partner who integrates your various cloud environments - a cloud "master chef" of sorts.

Your applications menu: the hardest to get right

The application layer is the layer where business processes are realized. This is where the user finally gets to interact with the IT environment. It is also the layer where the corporate strategy gets implemented in terms of various transaction systems, workflows, reporting systems and databases. Getting this layer wrong essentially means that your strategy may not get implemented well, or in time.

It is also the layer with minimum standardization across organizations (in some cases even within the organization). So any treatment of applications always ends up in a bespoke solution that is tailor-made for the client organization. The uniqueness of the applications layer presents a huge challenge when an organization is trying to "cloudify". There are no standard off-the-shelf replacements for a given application or a set of applications. This is quite different when compared to hardware layers, where it is easy to find an equivalent cloud based replacement for the MIPS and storage you require.

The application layer is also a key asset since business rules are embedded in the layer. Consequently, security and access considerations become even more important. All applications may not be equally amenable to move to the cloud. If only a part of them are moved, then the question of integrating cloud-based applications with non-cloud applications becomes all the more critical. The user experience has to be seamless as well. It should not matter where the application is actually residing.

The cloud applications approach: different flavors for different applications

Approach 1 - Identify applications where 'one recipe fits all': First, understand the application landscape from a business strategy point of view. Applications that do not create sufficient differentiation (which is the whole point of a business strategy) need to be looked at differently than applications that truly create the differentiation. The undifferentiated ones can be good candidates to move to the cloud. If we stretch the argument a bit more, one can actually move to prebuilt standard applications that realize the same business process over the cloud. As an organization, it may be better to reduce the complexity and reduce costs by moving to a more standard off-the-shelf prebuilt solution.

Approach 2 - Your horizontal applications are easier to cook on the cloud: You may also need to map applications that realize horizontal business processes (processes that any organization will have irrespective of the industry they belong to), and applications that realize vertical business processes. As a general rule, horizontal business processes can be moved to a cloud-based solution pretty easily. These include general ledger, accounts payable, accounts receivable, and human resources.

Approach 3 - Some of your applications belong in your on-premise menu: Evaluate existing applications for architectural flexibility that enables movement to the cloud. Given that the cloud is a relatively new paradigm, it is unlikely that there are many applications that have the flexibility.

Approach 4 - Your mission-critical applications may need a different recipe altogether: You may need a different approach for the truly critical, core applications that create the differentiation. By definition, these applications will always be customized for a given organization and will be treated as important assets that warrant a higher level of security. When such applications move to the cloud, the real value of the cloud really comes from the other layers such as the infrastructure layer. The other option for you is to evaluate a cloud-based business platforms that allow extensive customization. If these fit your needs, adopt the platform. Such platforms can bring your organization closer to getting the best of both worlds ― the best of the cloud environment, without sacrificing the unique value of the application itself.

Now serving: cloud-based business platforms on a platter

Infosys has taken a unique approach to help our clients to adopt the cloud paradigm. Our business platforms allow clients to create applications that are truly unique to them, but build and run on a cloud platform. Find out more at www.infosys.com/edge

The master chef: a cloud ecosystem integrator

In this role, Infosys enables clients to assess, design, implement a cloud a strategy ― without compromising the business value driven by the core applications that are unique to the organization. Delve deeper into what we do at www.infosys.com/cloud

Before you well and truly embrace the cloud paradigm, do remember that the application layer really needs to be thought through well. Depending on your appetite, we recommend a judicious mix of on-premise, private, public and hybrid cloud. And of course, find yourself a master chef to integrate your pan-cloud panoply. Sounds like an extremely successful cloud menu, doesn't it? A Michelin-starred one, maybe? Bon appétit!

 

September 11, 2012

The Big C in the Big D

 

Guest Post by

Timothy Sutherland, Senior Technology Architect, MFGADT Online, Infosys

There seems to be a lot of buzz surrounding the Big Data 'phenomenon' at present. While I suspect much of this is marketing hype and nearing the 'peak of inflated expectation' (driven perhaps by the same marketing folks that played a leading role in the dot.com boom crash at the turn of the millennia), clearly there is a genuine requirement to harness (or more specifically, exploit) the unfathomable volume of data that is being generated.

For me, Big Data is probably the convergence of mobile, cloud and social computing (Web 2.0), or perhaps more generally, an evolution of pervasive computing (but by no means a revolution). It was probably inevitable that this would become a hot topic, but that's easy said in hindsight.

While there is no clear-cut definition of Big Data, it is generally defined as data that is difficult to managed or analyse using traditional data management tools, methods and infrastructures. Clear as mud.

By definition then, surely the challenges that apply to managing this data could equally apply to content? Notwithstanding, data ('structured') has inherent semantic meaning whereas content ('unstructured') has intrinsic value and relevance in the context of the underlying the process that generated it, is consumed by it or interacts with it. The dimensions of volume, variety, velocity and veracity equally apply to unstructured content. Big Data is fuzzy and so is content.

Moreover, when we consider that content is becoming increasingly difficult to manage, even those organizations that do have the requisite maturity, capability and intent are struggling. It's spiralling out of control. Consider, generally 80% of information in an organisation is unstructured and the overwhelming majority is either not managed or poorly managed. Big Data in this sense is a much broader set of information across an enterprise and includes Big Content (Big C). Content remains King (and I am not just referring to web content).

Indeed, according to a recent survey by AIIM (see http://www.publictechnology.net/sector/central-gov/study-getting-big-data-out-digital-landfill), the majority of organisations feel that unstructured data is not exploited to the degree which structured data is. Many respondents to the survey were either unsure or felt that unstructured data was poorly used for analysis and decision making. How true.

Often, where the Big C is considered, and it is, it inevitably involves references to the huge datasets generated by the likes of Facebook, Twitter and Google. But of course Big Content is much more than that isn't it? So where does the Big C (Big Content) fit in the scheme of things?

Content management is more aligned to 'systems of engagement' rather than systems of records (see http://www.globallogic.com/geoffreymoore.html). Although in practice the distinction is perhaps not so clear-cut. Generally though, when we talk about Big C in the context of Big D, we are interested in both data AND content that is transacted during those interactions with a customer, employee or department.

How do we leverage this information? How also do we provide a unified and meaningful view between unstructured and structured data? How can we build bridges between the structured and unstructured repositories? Where can we exploit advances in search, semantics, categorisation and content analytics to leverage content and data?

While the leading content management vendors have extended their portfolio to include offerings to manage the lifecycle of a diverse range of new content types (such as blogs, wikis, tweets, digital assets, cases/contracts and other 'transactional' data) what appears missing is the meaningful analysis of this content.

Today it is possible to capture, manage, store, deliver and preserve content, but how does one leverage this content? How can content be used to improve analysis, decision making, performance and customer experience? How do we aggregate, consolidate and interpret the gargantuan volumes of information that is being generated? How do we sort the content that is of value to the organisation from content that is destined for disposal or the digital wasteland? Specifically, what role does content play in decision making and strategy? Or is it merely a by-product of the decision making process?

I believe the opportunities for Big Content are to extend the boundaries of content management beyond the document centric enclaves or silos we are accustomed to working with. That's because for a long time content management systems have been viewed as just that - a tool for managing the lifecycle (cradle to grave) of content, and specifically, documents or web pages.

What's needed is a fundamental change, a paradigm shift if you will, that can take the silos of information across an enterprise and combine this in novel or interesting ways. And how we think about content management systems must change. Content is an asset that must be leveraged and content management systems are platforms or enablers rather than packaged solutions.

The requirements is not just storage and management of Big C, but the analysis, re-use, standardisation (of content as much as process), interoperability and integration (structured and unstructured).

The Big C has an important place in Big D and the Big C should be given the attention it rightfully deserves. How the markets will address the Big C challenge is something we should all watch with some interest. Certainly new challenges will arise but new opportunities will quickly follow.

 

Guest Post by

Timothy Sutherland, Senior Technology Architect, MFGADT Online, Infosys

September 4, 2012

From 'just support' to 'highly strategic' - the evolution of application maintenance and operations (AMO)

"Albatross". "Necessary evil". "Budget eater". Over the years, application maintenance and operations (AMO) has received many undeserving monikers. Beyond the CIO's office, AMO is an enigma - misunderstood and written off as cumbersome and constant re-tooling done by "those IT geeks". This is a pain shared by most CIOs across leading global organizations. And during my many interactions with some of these CIOs, they have shared their angst - no management buy-in for AMO, the perception of being a cost-center, and worst of all, the inability to see value beyond "keeping the lights on".

But over the course of many engagements, I have witnessed application maintenance and related operations evolve and manifest in various avatars. AMO has played a key role in simplifying processes and driving enterprise agility. That is one of the reasons I'm writing this blog post - as a rejoinder, as a myth-buster. Because for every ill-conceived notion about AMO, there is an equally conceivable and readily discernable fact. So we're going to engage in some fact vs. fiction. Where fiction is a common-heard fallacy, and fact is the reality as seen through the eyes of CIOs and technology consultants.

Fiction: AMO is purely about optimizing cost and incremental efficiency improvements.
Fact: AMO is an unsung hero, and is delivering value in areas you may not even have thought of.

CIOs have typically used interventions such as vendor negotiations, vendor consolidation, outsourcing, and off-shoring. When they started out with these, cost and efficiency was the driving factor. But today, there's more to the story. AMO is capable of putting business before technology, and not the other way around. The right partner and the right approach can help ensure that each application delivers the desired business value to their users. Think about a merger or acquisition - the first, and often most important step to connect the two organizations is AMO. Think about how "those minor tweaks" to applications help make transactions faster and safer, saving time and effort for business users.

Fiction: Our business processes keep changing because our applications keep changing.
Fact: An application changes owing to a business decision, and this change will standardize operations for the better.

Today's enterprise IT application landscape is constantly in flux. With new technologies to be leveraged, it obviously makes sense that business users expect more functionalities from their applications. So take the business process view -- for most parts, the underlying business process itself does not change dramatically as long as the business itself does not change dramatically.

Many enterprises periodically embark on large-scale process harmonization and standardization projects. Usually they leverage an enterprise-class package such as an SAP or Oracle to drive the standardization. The landscape is supposed to be simplified with "one version of the enterprise-level template". Unfortunately such a simplified landscape exists only for a short while, as business starts demanding change, and over a period of time the application landscape becomes complex once more.

Fiction: IT consultants optimize application without understanding process linkages.
Fact: There are consultants who think process instead of application or functionality.

Since applications are only a means to realize an underlying business process, it is important to get a handle on the process itself and its linkages to other processes in the business. Most AMO teams start with a detailed inventory of processes and applications that support the processes. When a new partner is taking over, AMO is usually a great time to do this base-lining exercise. Infosys for instance, has a comprehensive set of tools and methods to create a business process inventory and an application inventory. This allows the proper mapping of process to application, with mapping of dependencies.

Fiction: It's better to have decentralized AMO operations, so that each LOB / function can customize the same application to their need.
Fact: AMO is easier to manage with a single global team.

By having a single global team to support AMO for a given set of business processes one can minimizes the changes. If one team takes a business process view and runs AMO, then the implications of changes requested versus the change itself can be easily understood. More often than not the change requested is not a must-have change and even if it is a must-have change, it is easier to control the change process by with a global AMO hub. And if a similar change has been carried out at a different location, the same team can do it faster - they know just what needs to be done. And think about what this team could do in terms of enterprise-level standardization.

Fiction: AMO teams don't understand business - their interactions should be limited to the CIO's office.
Fact: AMO teams are capable of strong change management control.

To get the best out of the AMO process to drive standardization, the AMO team needs to be empowered by both business and IT. This will avoid costly rework and failure in the landscape.

At Infosys, we've been using our Global Delivery Model to deliver on all the above facts. Going beyond "what's expected" to "what's possible". What do you think? Do you have more "fact" to counter all the "fiction"? I welcome your views.

September 3, 2012

The Search Engine Psyche Part 1

 

The Googles of the world have changed the way we searched for information forever. Today we can access vast oceans of knowledge with just one click. It changed the way we remember, categorize, search, and organize information. Today's generation want search functionality everywhere, right from their smartphones to the documents they are accessing are expected to have some basic search features. What happens when such employees enter organizations and are faced with an inefficient search which won't respond in the way which they have always taken for granted?

 search.png

 Web Search Process

 

Employees expect a friendly and intuitive search experience at the office as well, something similar to web search with which they have grown up. Here the Enterprise search comes into the picture.

Enterprise search is generally defined as "the practice of making content searchable to a defined audience, from multiple enterprise-type sources such as databases and intranets."  Enterprise search helps in locating the right information at the right time for the right people. If you have a wonderful knowledge repository through which employees can't navigate or search the information they are looking for, you have missed the point!

Listed here are a few Benefits of Enterprise search which can help your organization become much more efficient & responsive:

benefits.png

 

Informed and Smarter Decision making: Your employees will be able to effectively search and get enough data points to make an informed decision in a time efficient fashion and unlocking the hidden value of your information systems.

Intuitive Search-Better user experience: Enterprise search helps you provide a better user experience through superior design & user interfaces.

Guided navigation: Enterprise search helps you in providing better and 'contextualized' results to users through guided search. It provides users with filter criterions through which they can arrive at precise results without wasting time on unnecessary categories. This helps in providing the much needed context to the search capabilities.

Increased Efficiency: Your employees would be enabled to spend much more time on productive tasks rather than searching for information which is difficult to locate.

Global workforce enablement: People working in different time zones would be able to search efficiently the information/answers on the internal search rather than waiting for people to respond due to time zone differences saving you precious time!

Device agnostic approach: You can enable your employees to search information from anywhere using device of their choice. In these days of BYOD (Bring your own device to office) it is much more appealing to provide such a capability and independence to your employees.

Regulatory Compliance: Enterprise search helps organizations respond in a swift manner when faced with a legal issue, it helps in furnishing all the requisite documents quickly which otherwise would be difficult to locate from disparate locations.

So it makes great sense to invest a bit in an intuitive search system which helps your employees in getting the right information to make those important decisions. Provide a friendly search system, maybe something similar to a usual web search which will help your employees save a lot of time while fully utilizing the information assets present already in your organization. Happy searching! 

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter