BPM-EAI

Tipping Point for Enterprises towards Cloud...

2012-01-23T20:49:58Z

There is one objective for IT operations - reduce the existing operations cost and ensure a good sense of predictability on discretionary spending towards growth and scalability. All the initiatives whether it be SAAS, Public Cloud, Private Cloud, Virtualization are all aimed towards the same basic objectives. But enterprises still clamor for the right fit, fail in many of the initiatives and reducing the cost becomes a challenge. The IT vehicle hits the brakes on the road towards its supposed destination
It seems as if Cloud solutions could provide the mix of reducing the cost as well as provide tools and accelerators for new development. Sometimes it becomes a risk when the information transacted is part of the cloud and not “in” the Enterprise Infrastructure.
Some questions posed to the IT leaders are whether and when to go for cloud? Next is what all should go in the cloud? Should one have multiple clouds or single cloud? Should you have a mix of Private and Public clouds? and so on…
The theme of this write up is whether the “Enterprise Integration Services” could serve as the tipping point for the cloud and pave the way for exponential growth in usage of Cloud across the enterprise technologies.

]]> As mentioned in an earlier blog, just like a human being all enterprises are unique and they should not copy or get sold with an idea from a product vendor or a SI provider. Almost all enterprises have a multitude of applications such as SAP, Oracle databases, IBM Mainframes and integration technologies such as TIBCO, PEGA, Software AG serving the business needs. Most of these vendors also have their “Cloud Solutions” for their respective suite of products. Does it make sense for enterprises to go for a cloud solution for each of them or some of them? - As usual there is no one clear cut solution.

Consider a sample enterprise with a Production class having:

a. An SAP system for Back Office and Accounting Settlements.
b. 10 shared Oracle database servers supporting a wide variety of operations.
c. A Mainframe instance with having multiple modules of legacy calculations.
d. App Servers such as Weblogic, Websphere to support the UIs within the enterprise as well as customer oriented websites.
e. A BPM platform comprising of key business processes.
f. A tiered SOA layer comprising 100s of services that integrates between these applications and business processes.
g. Other ERP Systems for different functions such as CRM, SCM, HRM.

Putting monolithic systems like SAP, Individual ERP softwares, Oracle database in clouds is comparatively easier because of less customizations around such systems. But it will not save the bulk of the cost incurred by an enterprise. It is probably better off to keep them within or create a private cloud concept around the systems such as in SLAs for software life cycle. The objective is to faster time to market and reduction of overhead cost.
Glancing through the BPM and SOA components of the Architecture where it is all in house created services using either standard tools such as IBM, TIBCO, PEGA, Software AG and even .NET or Java based web services one needs to understand the depth and breath of such systems. Typically such software systems are like the virtual nervous system transmitting messages from one system to another in the body of the enterprise. Putting them in cloud is like getting the nervous system out of the body. Will it first of all work?
At the same time, it will also be the right candidate for any organization who is pursuing Cloud and wants to keep only the brain as part of the organization but all other parts outside of it.

In building Tomorrow’s Enterprise, organizations looking out for transformation in Information Technology should well think about first whether in-house mechanisms exist to get the same benefits as one gets with Cloud. Most of the time they will do and if there is a valid case to go for cloud, then integration services could well serve as the tipping point for enterprises to migrate towards a Cloud based Organization.

Evolutionary Social BPM Adoption Model and incorporating social features in existing BPM Solutions - Part 2

2012-01-18T10:12:41Z

In our last part, we were discussing the Social BPM Adoption Model, and we saw how collaborative process design is an integral part of the model. The next evolutionary step is to extend collaboration from process design to process execution through runtime participation. In this stage, though the process participants are fixed as in any conventional BPM solution, they are enabled with social tools that help them in better collaboration. These social tools are integrated into the BPMS landscape and can provide features like chat, VOIP enabled calls, commenting on tasks, providing ratings, voting mechanisms etc.

]]> A significant upgrade to run-time participation is enabling social collaboration by routing tasks in the process execution to process participants who were not envisaged to be a part of that process during deployment time. These tasks can be executed by participants in the social media and the process continues taking those inputs.

Extending existing BPM Solutions with Social Features

The Social BPM approach is oriented towards providing a method for process designing and deployment extended with Social interaction features. This essentially calls for extensions to the traditional BPM solution. The objective of process design is to produce two kinds of models viz. Social BPMN model (BPMN + social interaction features) and an Application model.

The Application model describes how the application behaves independently of the deployment platform. It is mostly used for high level verification of application features, prototyping and can also be used to generate application code. The social extensions for BPM User Interface includes design for collaboration as compared to task oriented User Interfaces. This includes UI components / widgets to facilitate chatting, sharing common spaces, enabling community interaction both internal as well as outside the corporate firewall, voting, rating etc.

The process frameworks and templates that are now typically restricted to packaging business rules, reports and workflows by specific industry domain or process would get extended to include components to enable social interaction. Accordingly the current availability of connectors, adaptors and services that are restricted to enterprise systems (like SAP, Oracle etc.) would be extended to include communities and social sites (connector for Facebook, Twitter etc.). Crawlers on the social community would provide valuable intelligence to enhance business processes.

Monitoring and Process Analysis would get enlarged in scope to cover social process monitoring and provide inputs to optimize business processes. This also provides a base for Social Intelligence

For those interested, you can find my detailed paper on how Social process design, execution and Social Intelligence can enhance Customer experience @ http://www.infosys.com/BPM-EAI/resource-center/Documents/social-design-execution-intelligence.pdf

Evolutionary Social BPM Adoption Model and incorporating social features in existing BPM Solutions - Part 1

2012-01-03T12:55:58Z

I was making a presentation to a CIO of a large bank headquartered in Netherlands on the need to have an evolving strategy around BPM as several enterprise spaces are morphing into one another and the space is constantly shape shifting. It was during that time that I realized the strong connect and the peer pressure that exists among the CXO community in Netherlands. The CIO of the bank was intently looking at "extending" their existing BPM solutions by incorporate social features because he came to know from one of the CIO of a competing bank that they had just then kick-started a similar program. I was happy because all we were hearing till then on Social BPM were from the analysts without significant corroboration from our clients on the ground. This set me thinking. I believe we are currently witnessing trigger events for an extensive social play in the BPM space. I wanted to share my perspective on the need for an evolving Social BPM adoption model and how various components of traditional BPM solution can be extended to incorporate social interaction features. Social BPM Adoption Model

Social features can be introduced at multiple levels while designing BPM solutions. It is pertinent to note that not all business processes would be "socially amenable" and the need for providing social features to BPM solutions must be driven by the overall process requirement.

At the lowest level, we have the conventional BPM solutions that are built top-down and are model based process execution solutions. They are characterised by rigid task allocation to different process participants with communication among process participants being restricted to task allocation and outcome communication and often facilitating Straight-Through-Processing with almost no communication between process participants. Tasks in the process are defined within boundaries, process participants are known before hand and pre-registered and workflows happen based on well-defined logic and escalation rules. The communication among different process participants are governed and controlled by user interfaces (either inbuilt within the BPM suite or on a separate UI layer), with outcomes, tasks and exceptions being communicated via email or task inboxes. Majority of the BPM solutions that are implemented today fall under this category.
The next step in Social BPM evolution is to enable collaborative process design. This throws open process design to multiple people who can collectively design business processes. BPM vendors like SoftwareAG (ARISalign) and Metastorm (M3) have come out with a collaborative platform for process design in the recent past. Collaborative process modeling provides a compelling value proposition if the BPM teams are geographically dispersed. This is the starting point for Social BPM in most cases.
The next evolutionary step is to extend collaboration from process design to process execution through runtime participation. In our next part, we will look into how existing BPM solutions can be extended with social features.

]]>

An Architecture for KPI - Most needed but ignored

2011-12-27T23:18:19Z

A branch manager is interested in seeing whether there has been abnormally low or high transactions in a particular day. A Banking regional manager is interested in knowing whether there has been an increase in customers with a recently launched campaign. The country head is more interested in knowing which areas have performed better or worse as compared to a predefined expectation. The global CFO is interested in seeing that if there is a revenue at risk due to any kind of external or internal impact. The HR head is interested in seeing the percentage change in attrition due to an initiative to increase the employee hours in the branches for providing better customer service. The CIO needs to look at the benefits of the discretionary spending over the last year. An IT manager overseeing a multi-million development program wants to estimate the value and even compare the actuals once it is delivered and is ready for measurement.

Most of the information in such examples conventionally have been available as offline reports run through batch jobs scheduled as per the need and convenience or through different other means.

The question is how should an enterprise design its Information Architecture to deliver the right set of KPIs and also make it more usable at all levels. Here are some tips:

]]> From the top of an organization to middle to the lowest management layer, Key Performance Indicators are necessary for efficiency and effectiveness for the managers no matter what the business that organization is in whether it is retail, banking, telecom or energy.

Create a separate roadmap for KPIs. If need it can be part of Business Intelligence or Centralized corporate reporting. It can also be merged with the reporting, compliance requirement for public companies to report corporate actions and other information as a listed company.

Capture the KPIs for every user in the company with Purpose, audience and usability. To assess the usability, simple question to be asked is “What is the impact to the user if the KPI is not available?” Prioritize these requirements with applying multi dimensional criteria as relevant to the business.

Capture the timing of the KPIs whether real-time, daily, monthly, weekly. This will define the tool sets and cost of investment for implementation.

Create the conceptual architecture using the Organization chart as the base template. Map each row of low-level, mid-level, high-level, executive-level, CXO-level KPI requirements.

Create the blue-print for executing and developing as a single program spawned into multiple projects.

Interestingly this is how and where CEP, Predictive analytics, Just in time Reports and other state-of-art tools to get the right information at the right time can be easily fitted in the enterprise instead of force-fitting for tactical benefits.

Clouds in my coffee!

2011-12-21T11:22:01Z

In the place where I live on this planet, monsoons come twice. However, in the virtual world of IT services, it has been raining all through the year, due to the “clouds”. The concept of “cloud” based provisioning is not new to mankind. During the Agricultural civilizations - clouds from vapors of water ensured that consumption needs of the humans are met. During digital civilization - “clouds” are provisioning digital consumption needs of people today.

The foundation for contemporary cloud based computing takes us back to early 20th century where the world was witnessing the “first war of the cloud” between Tesla and Edison- for the best model of transmitting “electric power” in a “cloud” model [Grid-based-AC vs. Short-distance-DC]. At the same time, in Cambridge UK, Alan Turing, proposed a model of computing, later called ‘Turing Machine’ which has infinite memory [tape], infinite computation capability. Grid based supply of electric current has always been compared with the Cloud-based computing facility.

]]> In geometry, we can never realize a representation of a straight line between two points in plane. The moment a pencil/pen touches paper to show case a geometric point - the abstraction is lost and a point now can be measured. Technically, a geometric point’s size cannot be measured, as it infinitely small. The point made from ink/carbon on paper is measurable - even if it is nanometers, it is still a quantifiable size.

Likewise, we can never realize the theoretical Turing Machine in form of silicon chips or flash memory - as all these technologies are bound by physical limits. All the efforts today, with whatever brand name they are being provided viz. Cloud, Big Data, SOA, Utility computing, Data-Centers etc. are all little steps towards making the Turing machine a reality.

In last one year, there have been too many announcements by Software Vendors that they are now Cloud-enabled. One can hear terms like Infrastructure on Cloud, Application on Cloud, Integration on Cloud, BPM on Cloud, etc.

There are certain applications that can never be Cloud-ized - like employees’ SSN details, salary details, bank account details, core applications. Purists would stand up and say, there is a “private cloud”. The moment a “cloud” is “private” then it is not a cloud as in “the cloud”. This is akin buying your own generator for your electric power needs and calling it as “private grid”, as against using electricity from the “grid”. It might just sound like adding floral drape to a bat suite.

Techies out there!! Let’s remember the KISS principle! “Keep It Simple and Silly!” and not get drawn into the humdrum of publicity, with which the world is gradually losing on true purpose of IT in an Enterprise. Hence, the title of this writing is taken from the soulful 1972 song of Carly Simon’s “You’re so vain!”

[Editor’s note: The opinions expressed in this blog are no way connected to Infosys or its subsidaries’ view point on Cloud or allied technologies and are purely the authors.]

Global Integration Projects: Challenges and Opportunities

2011-12-15T08:48:08Z

By Paresh Deshpande

Over the past couple of years, I had opportunity to work as integration adviser for global IT projects of an enterprise. Watching closely execution of such integration projects was a great learning experience. I couldn't resist sharing my experience from one such project I worked on recently.

]]> This particular project involved replacement of existing IT solution with standard SAP solution. IT solution being replaced was used by business users from around the globe. I was responsible for managing the integration aspects of this particular project.
In terms of challenges from this global integration project, my observations could be summarized as follows:

a. Time zones
Global IT application being replaced was receiving data from regional / country level systems from around the globe. Integration team had to address different time zones in which development teams of local feeding systems operated. This led to constantly changing hours when integration team had to interact with the relevant colleagues managing regional / local feeding applications.

b. Multiple IT services vendors
Each of the geographic regions and countries had preferred IT services vendors. Integration touch-points accordingly involved interaction with different vendors for each of the interface connectivity needs.

c. Languages
Language related issues heavily influenced execution of the integration activities. Language related dependencies included were translators joining conference calls with IT departments from Brazil and Mexico.

d. Parallel Integration Support
Due to phased country by country rollout of the SAP based solution, integration support for both old and new IT applications had to co-exist. This presented intricacies in terms of number of interactions which integration team had to manage as part of the switch-over activities.

e. Changing Integration Landscape
In the same time that the IT solution was getting replaced, integration landscape of the enterprise was also undergoing changes.
Overall, I was quite amazed by the level of complexity one has to go through for an integration project of such a global scope. Replacing an IT solution while its middleware is also getting replaced or phased out is a daunting task in projects of such global nature.

While such challenges exist for global integration projects, there also exist several opportunities to improve the execution and better serve the business. For this particular project, I especially benefitted from the goodies in the form of Infosys best practices when converting these challenges into opportunities.
a. There is huge scope for minimizing the number of integration technologies used for such global applications. Infosys EAI best practices for integration design such as Use of Common Messaging Model and service based approach not only help simplify operational complexities but also make the integration future-proof.

b. For a global organization, optimization of number of integration landscapes (for e.g. regional and country level middleware systems) would help minimize number of areas where the integration activities are restricted to. For example, I facilitated merging of regional middleware systems during execution of this project thus optimizing the integration landscapes.

c. Program office which aligns different IT programs (integration IT program, IT solution replacement program etc.) plays crucial role in timely and cost effective execution of such global projects.

d. Another area which could help execution of integration projects is reducing the number of IT services vendors that the enterprise uses. Opting for single vendor such as Infosys is for example ideal scenario for such integration projects.

In hind-sight, projects involving global scope have to be planned in a way such that some of the key aspects of global integration (for e.g. time zones, language dependencies, and overlapping programs) are assessed upfront. This would help execute the global projects at individual country / region level effectively.

ICC: Standing the test of time

2011-12-01T09:46:53Z

“Truth is stranger than fiction, but it is because Fiction is obliged to stick to possibilities; Truth isn’t.” said Mark Twain. In our day to day lives, we come across many truths which we blissfully ignore and take it as a matter of fact. Gravity is one such “truth”, while many a pseudo truths have been falsified as stated by Nicolas Taleb in “Black Swan”; Sir Karl Pooper went on to say about scientific truth are of two types, first type - which is falsified and the other type - waiting to be falsified.

Notwithstanding the debate about the truth and absolute truth, there are certain aspects which remain true even after many centuries or many ages that have passed by. “Gravity” and “Division by Zero” are two prominent examples that come to our minds when we think of Universal Truth. The most modern technologies, when applied cannot falsify these truths. May be a human/animal can feel a sense of Zero-Gravity for a moment or a few, but again they are subjected to the ever present reality.

]]> Why I am talking about Universal truths and their falsification on a blog about technology. It is because of a conversation I have had with a colleague of mine in recent past. We were discussing about the approach one needs to take for setting up of an ICC [Integration Competency Center] for an Enterprise and the logical steps in the route. While we were sharing our experiences, suddenly, to my surprise, he mentioned, “Don’t you think what you are talking is a decade old concept?!” and went on to say, “Why are we even talking, if all we are to do is get the canonical models up, have a single source of truth {truth is a fancy term in IT world, used to address data}. We have gone through that route many a times, tell me something which is relevant to today’s technology.”

I could not believe that such a senior EAI practitioner was talking about “relevant” changes in the approach to setting up of an ICC. For once, I wanted to ask him, does The Third Law of Motion proposed by Sir. Isaac Newton, hold good today? - As there has been a tremendous leap of technology since last 300 odd years, since it has been proposed. {Now, I do not want to get into a debate here about “classic” Physics and “Modern” Physics.}

Agreed, that the principles of ICC are decade old, is not RDBMS a three decade old model. That very concept on which Oracle Corporation has made millions! May be with slight changes to the access to the table data and some niceties, Oracle database at its core still is an RDBMS. So is that case with all the networking protocols, except for some make-ups here and there, the TCP/IP model proposed in early ’60s is still at the core of the network.

Why are these two examples relevant even today - there no equivalent or better model available for data-base management today, nor for the networks. Even if there have been some proposals, they are still at the research labs and or not yet made commercially viable for wide spread acceptance. The practical realization of a true Turning Machine (with infinite memory, infinite tape and infinite processing capacity) is yet to come.

Till the time, some better approaches are made available; we are to live with this decade old, proven “commodity” approach for setting up an ICC.

Managing NFRs and Predicting Performance

2011-11-10T23:08:25Z

In one of the past engagements, an SOA solution involving around 80 services being defined did not have any documented NFRs. When we requested to the business team, the response was quite timely - that all services need to respond within one second, all should be highly available, all should be available in multiple data centers and should be part of disaster recovery.
There is nothing incorrect in having such a requirement but it is neither realistic nor beneficial.

In another case, a programme going live in another few weeks was doing its load testing as the last phase of testing but most of the test cases were failing and not meeting the NFRs putting a risk to the programme.

Questions that naturally arise - How should one manage the NFRs in an enterprise without getting into ‘obvious’ troubles? How can we use the NFRs to predict the performance even before the solution is in build or even design phase?, What-if, after doing this exercise and the resultant prediction itself is leading to unexpected values?, Can we chart out a plan to get to the end goal with right justification to the cost?

These are exactly the questions that need to be answered through Performance Engineering.

]]> Most and if I could take the liberty to say ‘All’ enterprises face this key challenge in their IT projects - getting the ‘right’ performance from their solution whether it be response time, processing high transaction volume, predicting the response delay in case more number of users are using a system, time taken to failover, data center switch and other aspects of availability.

Besides it’s about not just meeting the NFRs in load testing and doing well in initial days after implementation but the solution should sustain itself over a long lifespan.

There are some simple steps to manage, govern and address the NFRs.

I. Get the requirements as top to bottom approach:

i.- List out the Services as part of a business flow according to their gradation compliant to the Enterprise SOA standards - whether they fall under business services, technical services, foundation, mediation etc. For e.g. for an Auto insurance company a Claims Process will require services such as ‘getCustomerData’, ‘getPrimaryInsuredInfo’, ‘getVehicleInfo’. Group these together.
ii.- List out the requirements for each of this service - lets say 1s response time as expected by business.

II. Put in Granular details (helps in capacity planning and predicting resource utilization)

i.- At what times the services need to be available?
ii.- What are the usual peak times when the services will be hit with high volume? - for e.g. middle of the day for a Investment/Brokerage firm.
iii.- Are they real-time or scheduled?
iv.- Volume of Information being transacted; Volume of Users, Size in Bytes of each transaction.

III. Put in the ‘brakes’ for these services. (this helps in putting the negative clauses to help predict a closer to accurate picture)

i.- Data Access services has a dependency on the speed of database.
ii.- Assume a factor for parallel processing, which will give the throttling criteria.
iii.- Assume a factor for deterioration in a shared environment.
iv.- Network Latency. This will have two kinds basically, one for Intra datacenters or same geographic location and other when systems are separated such as in different continents and countries.
v.- Any other breaks, bottlenecks that is applicable to the enterprise.

IV. Standard Measures to the Environment

i.- Assume a standard environment based on existing Infrastructure footprint that these services are expected to host.
ii.- Include all parameters such as Memory Utilization, CPU speed, Server Capacity, no. of services that will go in each server including load balancing instances and stand-by instances of the services.

Based on the above parameters, the response time as well as utilization can be easily predicted including the cost of Infrastructure. An Excel workbook with simple formulae can publish a report with predicted response times and whether it matches the original NFRs.

Now comes the second part where diagnosis is required for the NFRs that are not being met. There are two options:

i.- Apply Tuning measures (theoretically, design change, scaling method) and change the values for environment and keep on doing till you reach the desired NFRs. - This will also help in predicting the Increase in Cost
ii.- Another discussion with the business where Cost vs Benefit analysis becomes the point of discussion to relax the original NFRS. Trade-Off done to benefit everybody. For DR there is another write up…

The same approach was used for the particular engagement and resulted in more than 50% of NFRs being relaxed to more realistic numbers and only 10% was identified for performance tuning and there were no changes required after load testing.

Why Private Clouds are not the most ideal choice ...

2011-10-31T01:11:35Z

It's usually not the best reader retention technique to give out the "punch" of the post in the title, but then there are some messages that are best delivered straight up.

Well, in my opinion, the debate of public cloud versus private cloud as the preferred starting point of an Enterprise's computing platform transformation is as old as the buy versus build choice. Having seen several large programs struggling with delivering complex application using open source coding platforms such as JEE, I am convinced that it's often a better idea to buy a product and customize it rather than try building it from the scratch.

]]> I visualise it in terms of a custom-built car for a celebrity. The approach involves choosing a reliable vehicle which meets the base requirements and then enhancing it with customizations that are required. While the design team might still struggle with trying to figure where to fit the bazooka or the bath tub (depending on the celebrity), at no point are they supposed to spend time on trying to build an engine that takes in fuel and converts it to kinetic energy to make the wheels turn.

If we extend the situation to developing an enterprise system on the cloud, it becomes apparent that the choices are similar. There are many patterns, or categories, in the world of cloud computing that you can use to meet the needs of your enterprise architecture. Some solve specific problems, such as security-as-a-service or testing-as-a-service, and some provide complete platforms, such as platform-as-a-service or infrastructure-as-a-service. They all have trade-offs and different problems that each solves. However, you must consider them all in light of your architecture.

In many organisations, there exists, what I call, the "Inertia of Control" against letting go of control of a platform to an external vendor. There are various reason cited for going for an in-house version of virtualised solution called "Private Cloud" such as security, legislation and fitment, however none of them really convince me that this is a prudent choice. In fact I believe there are a few very compelling reasons as to why a "Private Cloud" is not a good pattern to start the journey.

1. Technology Platform was never the problem.

When an organisation builds an enterprise system, the underlying technology is usually the least of their problems. While, most organisations may either see Red or Blue on this issue, I believe that its safe to assume that both Oracle and IBM (SAP as well) have reasonably well developed solutions available to meet most of the basic requirements. The problem has never been that they have had a physical operating platform as opposed to a virtualised platform which allows you to better utilise the underlying hardware. The problem was that the total cost of ownership (TCO) of the entire platform over the life cycle of the application was very high. Right from the procurement and licensing to the actual customization effort and the maintenance and enhancement of the platform was very demanding in terms of cost, effort and know-how.

Cloud Computing, with its inherent support for Pay-as-you-Go commercial model and On-Demand availability takes away all the hassle and management overheads related to maintenance and capacity planning. With a private cloud, the organisation still needs to have huge teams maintaining the underlying infrastructure as well as running the risk of ending up with idle capacity in case of erratic demand.

2. Best Practices bring in Agility

One of the other major advantages of the P-a-a-S solution is that it allows an organisation to purchase and industry best practice out of the box. Years of change management workshops fail to meet up to impact that can be achieved by implementing an Out- of- the- Box- In- the - Cloud solution. A Private cloud development program is always run to meet the constraints of the Enterprise and in the end resembling the traditional enterprise system albeit on a fancy new platform. Users, across the world, resist change, but getting them to move onto a platform that welcomes change and indeed manages change in a evolutionary fashion as opposed to a revolution does bring in the agility into the organisation to respond to changes. This means that when the business circumstances change, the IT system allows the head room for innovation.

3. Every Change starts with a Leap of Faith

According to David Linthicum's Cloud Computing blog , "The choice for adopting a public cloud requires that the organisation trust somebody else to deal with your infrastructure, applications, and development requirements. It might also mean that the IT leadership may need to shut down a data center or two and reduce the size of their kingdom. Many in IT seem to have a real problem with doing that; when asked to move to the cloud, they choose the private option each and every time.

What most decision makers need to understand that the choice to go for a Cloud based solution pattern needs to be driven only by business requirements and not by IT apprehensions. It also needs to be recognised that most of the major suppliers developing large Cloud platforms have spent billions of dollars in building a safe, secure and reliable platform and hence do deserve the benefit of doubt. Sure there are fall-outs like the Sony PSN and Amazon E2C debacle, but I believe these systems are at least as robust as any on-premise solution.

LIVE Update from TUCON 2011 - Part : 2

2011-09-29T01:55:41Z

While I talked about the theme elements and the key focus areas, Infosys participation at TUCON 2011 provided us with interesting insights to customer perspectives. More and more customers are finally looking at the Cloud seriously, something substantiated by TIBCO themselves.
]]> Some interesting announcements coming from TIBCO during the course of the event:
- Focus on zero latency devices and applications is taken to a new level in line with the philosophy 'Big Data wont help you in 6 months'. Particularly interesting is the concept that 'A little bit of right information before hand, is better than all the information 6 months from now'
- TIBCO web messaging platform to service mobile and web enabled platforms as an extension to TIBCO Enterprise Messaging Service
- SpotFire 4.0, that customers and system integrators are awaiting with equal anticipation after the eye-grabbing releases of the previous versions
- A wide multi-flavored TIBCO Cloud platform that promises to be a 'one click' deployment off-premise of infrastructure and software that would take weeks if not months to deploy conventionally.
- Tighter integration of newer acquired solutions like the Nimbus owned process discovery capability and Loyalty management solutions, with Context capture and collaboration capabilities. If the prototype that we saw is an indication, we are in for some very cool technology innovations
All in all, TIBCO's data handling capabilities and the focus on low latency, data context and extending the integration capabilities is likely to create a niche platform, that isn't probably covered in existing analyst quadrants.

While at the event, we had strong cross-industry participation at the Infosys hosted cocktail event attended over 50+ clients and prospects. This coupled with the insightful interactions in the solution showcase area, does provide a point of view around TIBCO across verticals. The BestBuy Europe and Infosys hosted session on Customer Experience challenges and the hosted solution deployed was well received too.

The highlight of the Infosys participation was the key note session by Ashok Vemuri on how as businesses move away from a product-centric model towards co-creation of value, tomorrow's enterprises will be defined by their ability to create and nurture an ecosystem of innovators. We look forward to sharing the video snippets of the keynote with you soon.

LIVE Update from TUCON 2011 - Part : 1

2011-09-29T01:13:03Z

In my pre-event blogs,I was wondering about the single overarching theme that TUCON 2011 sessions would center around. As the diverse range of keynote sessions unfolded with topics as wide as Innovation, Customer Loyalty, Agility, Mobility, the one standout TIBCO theme that was evident was - Data, more precisely handling data in the 21st century.

As Vivek and team put it at the start of the conference, it is becoming increasingly about "Right DATA, at the right time, at the right place and with the right context". TIBCO's strategy, value proposition and the roadmap of the product line-up is very appropriately centered on this notion.

]]> There were some interesting messages in the keynote, that alluded to the various facets of handling data in the 21st century, here's a few:
- MGM Casinos and Resorts highlighting the shift of the competitive battleground to 'Customer', and how they plan to leverage TIBCO's data handling capabilities for creating an intimate relationship with the customers, NOW!.
- Cognizant's session pointing to an interesting phenomenon, dissonance between the consumer experiences with technology vs the corporate experience with technology. Most enterprises are facing this when the employee who is uber-connected at home in the evening, comes in to work in the morning and expects the same ecosystem.
- Fedex CIO mentioning a key tenet that the logistics industry and enterprise IT at large, should find very valuable -"The info about the package is as important as the package itself"

Also, as much as it is about data at the right place with the right attributes, increasingly enterprises are moving away from control and command while handling this data. As TIBCO put it, Tomorrow's Enterprise functioning will be more akin to jazz that symbolizes improvisation, than stand for Common and Control for which Corporations were created in the first place, interesting thought... But very relevant, given the channels, types and size of data that the enterprise 3.0 is bombarding us with. Stay plugged in for more updates from the event, and keep the comments coming..

The Innovation Advantage: An Infosys POV

2011-09-27T15:05:44Z

Before wireless radio transmission was invented, long distance communication was a matter of days, months or years. But in 1895, a brilliant Serbian-Croatian inventor Nikola Tesla demonstrated a simple wireless system which could communicate over a 50 mile distance almost instantly. But he failed to commercialize the idea. Years later, an Italian inventor Guglielmo Marconi applied for a patent for radio transmission in 1893 and became the first man to commercialize radio. Marconi was later known as the inventor of radio.

History is replete with such inventions which never saw the light of the day until a market-savy individual understands its potential. Hindered by the slow spread of information and uncertain financing, a stereotypical eighteenth century inventor toiling for months in his underlit laboratory might achieve too little or too late.

The 21st century is indeed the best time to be an inventor. The vast repositories of information at one’s finger tips and viral nature of many social networking mediums ensure that many obstacles that inventors of the past faced are no longer present.

But does this mean that enterprises are thriving with innovation?

]]> The reality seems to be far from it. In a survey conducted in 2010, 95% of the 300 Fortune 1000 executives surveyed recognized the criticality of innovation to future growth; more than half of those surveyed admitted that their company had no tools or processes for fostering enterprise innovation and at least a third saw their absence as a barrier to innovation.

Paradoxically, the hunger for innovation among consumers seems to be higher than ever before. Popular websites like Facebook and Google launch wide-sweeping changes overnight in a race to please its online visitors with ever-increasing functionality. Consumer electronics manufacturers carefully plan their release calendars for several years in advance to intentionally stagger their innovative features. Retailers need to add the redundant ‘New and Improved’ phrase in most products to keep the consumer excited. To today’s impatient generation, if a company has not innovated for a while, it is as good as dead.

With the ever growing importance of fostering innovation within the enterprise; there is an increasing need to create more springs of innovative ideas, connect them with changing customer needs and allow them to shape the future organization.

According to Infosys there are three key enablers to build an environment of innovation:

Adaptability - Often disruptive changes within an industry gives an opportunity to reexamine the core strengths of a company. Insights from such an exercise might not just help reinvent itself but also ensure its longetivity.
Collaboration - Ideas spread faster than ever in in today’s socially networked world. Companies thriving on innovation have successfully created Innovation Platforms on a global scale to accelerate ideation and teamwork.
Simplicity - Over the year, organizations have grown through several mergers and acquisitions, resulting in process complexity and legacy systems. Perhaps, hidden under those layers of obfuscation is a simple and easy to understand business system. Simplification invites innovators to think and improve your businesses.

This Infosys point of view is further explored in the whitepaper ‘The Innovation Advantage: Build the Future-proof Enterprise’ and available for free download.

TUCON 2011 - More on BIAN as Core Banking Surround

2011-09-22T08:49:30Z

In Continuation with my previous blog post http://www.infosysblogs.com/bpm-eai/2011/09/tucon_2011_-_core_banking_surr.html#more on BIAN indentifing 250 service domains across Sales & Services, Reference Data, Operation & Execution, Analytics & Risk and Business Support business areas, I think it is an important event within Banking Industry in terms of taking a collective step including Product Vendors & SI vendor to bring in standardization which was really due for long.

]]> So how does this standardization help banks internally and collaborate with their partner eco system?

From a bank's perspective, it provides a great opportunity to adopt BIAN best practice of business area classification for its IT systems to reduce overlaps, identify opportunity for consolidation, establish the service catalog and lastly look at business area based legacy modernization

From a Product vendor perspective BIAN provides the basic guidelines of componentizing the application landscape and expose it as a re-usable service aligning well within the banking industry.

For a System Integrator partner, BIAN provides a unique opportunity to work alongside with customers and other industry players to jointly define, build and implement next-generation banking platforms based on a common architectural framework. To leverage this opportunity, Infosys has been enrolled as the newest member of BIAN comminity . Read here for more details.

At TUCON 2011, we will be showcasing our capabilities on BIAN and would like to discussion in details with you. Looking forward to meeting you to discuss more on the implementation.

TUCON 2011 - Exploring 'Big Ideas' in enterprise software: Customer Loyalty in Utilities

2011-09-22T06:15:53Z

In my previous blog I talked about one of the key 'Big Ideas' at TUCON 2011, - Customer Loyalty - , its implications in retail and some solutions and sessions that attendees could look forward to. A more interesting case of customer loyalty is being witnessed in an industry hitherto known to take its customers for granted - Electric Utilities.

A little over a year ago, customers of a leading west coast utility in US resorted to a major backlash following installation of smart meters that triggered fears of electricity bill spikes. Such was the impact of the "Bakersfield Effect", as it is called now, that a state sponsored inquiry was hinted and industry analysts talked about obstacles to Obama administration's $8 Bn "Smart Grid" initiative. An innocuous rate increase bill that didn't grab any attention when approved by the Utilities Commissions earlier, might never be the same again. And at the heart of the issue is - Customer Loyalty. The aspect of Customer Loyalty that is central here is Customer Engagement and co-operation with a service provider. My colleague Larry Rubenacker talks about this effect and its implications in greater detail in his blog on 'The Smart Customer Experience Mandate'.

]]> The Mobile, Agile, Social and 'Smart' utility customer expects more from her utility than a website to pay monthly bills! Utilities across the US and around the world are now looking at ways to engage the customers more and more, even if they are functioning in a regulated market with little choice to customers for switching their providers. This can only help the Utility in smoother change management of major initiatives like the Smart Grid, Energy Savings programs, Distributed Generation and other novel 21st century solutions.

A few months ago, TIBCO themselves made a welcome surprise acquisition of Loyalty Labs, a company specializing in Customer Loyalty offerings to retailers, hospitality chains and the like. I wonder if leveraging that for Utilities was ever thought about, but it would be very interesting to see how a platform like that could be used for marketing heavy applications in Utilities. At Infosys we have seen a lot of interest in applications like our Demand Side Management solution, that has been designed to build successful and planet friendly customer reward programs for Utilities. Drop by our booth at TUCON 2011 for a detailed conversation about the solution, or write back what you think about this new Customer Loyalty paradigm in Energy and Utilities.

Finally consulting think-tanks like ours are looking at devising Utilities centric solutions for driving customer engagement, customer loyalty and even customer retention where needed, using modern day platforms. Read more about Customer Service focused Utilities solutions here or interact with our industry experts sharing their thoughts on our Smart Utilities.

SOA Evangelist

2011-09-22T04:58:53Z

A football coach when he joins a school, primary if you may want to call it, has a herculean task of training the kids for the real game. Kids know that the ball needs to be kicked into a goal and more often than not, in their enthusiasm of game, kick the ball into their own goal; defeating the very purpose of match. There might be “dares” too, as to who would dribble the ball for most part of the game - which in effect stops the ball being passed on to an untagged player near to the opponent goal, missing out on an opportunity to win.

]]> Similar would the situation of an SOA Evangelist visits a client’s IT department for the first time. In the enthusiasm to go the next stage of IT maturity teams from different Lines-of-Business would want to reach the targets individually, like the a kid hitting self-goal for the notion of hitting a goal, irrespective of whose goal the ball is reaching. This is where the SOA Evangelist needs to play the role of a “coach” and get the buy-in from management to create “rules of the game”. This would help in setting the “goals”. This would imply in forming of a centralized SOA Center of Excellence, which has oversight across all the LOB IT divisions and the owner of such Center of Excellence would directly report into CIO of the organization.

The IT divisions of LOBs “dare” each other as to who holds the “ball” for longer time, by owning the services of their LOB for use within their Unit as against making them “Enterprise wide” Services. Again, as a coach gives directions to children’s team to “pass” the ball for team’s benefit, the Evangelist needs to conduct workshop and ensure that all the LOBs concur with the idea of “centralizing” services, for sake of overall benefit of the Enterprise.

The critical aspect for any coach is the state of the playground. If the practice field is not adequate or in state of despair, our coach needs to get funds/grants from school board, so as to provide a decent (if not state of the art) play ground with a good gymnasium which can help his students to be in fit shape. By the same token, SOA Evangelist too needs to do an assessment of the existing applications in the Enterprise’s IT landscape. Based on the outcome of such an assessment, the recommendations need to be discussed with CFO and CIO for seeking budgets. As a playground needs to be leveled and some of the sand mounds need to scrapped off, some of the applications need to retired (if the case be) or modernized (based on the stage of application life cycle) or replaced (if there are better alternatives available). Once the playground is leveled, a new turf would be laid, likewise a new application that can act as an ESB needs to be procured. As the budding players need good gymnasium, the IT hands of the Enterprise need to have good sandbox where they can work out on different approaches to “Service enable” their applications.

The school board looks forward to getting some results from the new formed team and coach, especially after spending to modernize the playground and buying equipment and other peripherals. A coach would plan a “victory map” for his school’s team like playing with a neighboring school, then playing at county level, then playing at state-level school meet and finally the national school meet.

As the coach starts paving way for the national cup with small local wins, SOA Evangelist too needs to start out with showing “results” small wins, starting with service-enabling critical applications at LOB Level, automating some of the processes based on these services’ orchestration, then service-enabling all the applications in all the LOBs and finally establishing a futuristic SOA 2.0 IT Landscape.