Containers have now been a new kid on block for Digital revolution
Containers are supposed to be one stop for moving and managing legacy applications. Containers now can deploy apps with different OS, different application servers etc. Though it has been good 5 plus years since Containers surfaced we see limited use of them.
Containers are still prevalent in applications area where companies are using them for quick deployment of applications. Use cases of containers in area of data and them replacing VM need to be weighed in as Containers do not offer same level of security or run multiple applications as a VM
My team is working on
- Container factory which easily moves applications and hosts it
- Framework to help programs which can use Container based on parameters provided
There are several threads from Docker, VMware . Looking for your responses , thoughts on this topic
Lot of Companies want to jump on to the Digital Transformation without understanding what is the need for transformation and where they need to transform. Above and over that Transformation should not be a technology driven initiative. Before kicking of Transformation teams should be clear on 3 main criteria
- What should we Transform
- Customer Experience
- Business Process
- Operational Process
- Where should we start Transformation
- Transforming Customer experience can have improving customer touch points, customer analytics, CRM systems
- Transforming Business Process involves Digital Modification, providing multi channel business
- Transforming Operational process involves worker collaboration, process digitization
- How should we Transform
- Use of Social, Media, Analytics, Cloud and IOT
- Use of Automation, AI / ML…
- List goes on based on the need
The table below should provide a clear direction on Digital Transformation
DevOps has been the buzz for last few years. Every CIO wants to have DevOps to implemented and so are companies who are eager to showcase their DevOps product suite. Several companies are not sure if they are DevOps ready or if they need to worry about implementing DevOps. This blog dwells on few areas which companies should evaluate before deciding to go in for DevOps.
DevOps implementation, usage depends on several areas
Organization Behavior, Team culture and basic tenets of DevOps such as Continuous Integration, Environment, Release Management, Testing, Data and Operations. The excel of DevOps can be downloaded by clicking on DevOps
This Map gives simple idea should I or Should I not for DevOps
One of the website for good DevOps reading is https://sdarchitect.wordpress.com/understanding-devops/
In recent days there has been a paradigm shift in Business Analytics with Search engines, Hadoop, Big Data and so on…What happened to BI solutions we used. Why did HP by Autonomy or Oracle buy Endeca?. Why did Oracle announce appliance using Hadoop?
DW professionals have grown up with Kimball and Inom approach for DW solutions. Product vendors started with tools which can help in Dimensional modeling, creating data marts, data warehouses. With data growing Companies spruced their hardware and came up with hardware based solutions such as Neteeza, Greenplum and Exadata. All these hardware solutions can handle huge workloads and make things move much faster. With social networking, explosion of mobile the data is not any more in a single format. Now the data can be a an audio file, e-mail, document or image. Traditional Data warehousing design may not be helpful in analysis of available data. That;s where industries are looking at different approach in mining these data and provide Users with result set in seconds.
Unstructured data has made Business Analytic s to rely on different technologies, hardware, data structure and search engines which can save and provide the information in a way which makes analysis easier. Companies have started building platforms and solutions to address the above need using Hadoop for fast analysis of unstructured and structured data.
Hadoop is a open source framework for running distributed applications to process large amount of data by allowing users to write parallel code. Can scale linearly with addition of hardware and gracefully handle failures in hardware. Hadoop was a sub / sub project of Lucene. This blog will continue with samples and usages of Hadoop…
Over past two decades several of Indian IT companies had a good rise in their fortunes. They inched closer to Big 5 consulting companies. All these happened because of several factors
- Low cost alternative
- Global presence in several countries
- Offering 24×7 services
- Better and bigger talent pool
While all these helped Indian IT services to grow faster than brick and mortar business anywhere. Along with this growth came some problems such as
- Higher salary
- Lower productivity due to attrition
- No Innovation
- No / minimum investment on newer technologies
- Management focused on cost savings
All the above has caused profits of some of the Companies such as Sonata, Mindtree to drop. This trend might continue for the other tier 2/3 players if they focus on openration costs as way to survive.
IT service providers need to revisit their strategy, focus on innovation and new technologies as
- Cloud becoming dominant in SaaS, PaaS and IaaS most the IT services players will feel the pinch. The SLA’s will be managed by cloud vendors and this might cause loss of revenue in managed services or reworking on cost for IT service providers.
- Innovation such as iCloud, Google sync will draw customers to look at a new paradigm where Network, System and database administration are more so managed by Cloud players than IT service providers
- With less investment in newer technologies and more focus on cost savings the management will have staff with no knowledge of latest trends and happenings in the Industry. This might lead to retrenchment of work force.
Gone or the days where IT services will develop large applications. Force.com, Amazon and Google Apps have applications which are ready and can be customized to needs. There are thousands of these applications which can be used as is. IT service providers bubble is sure to burst if they do not think of innovation, Cloud as path for transformation to their clients.
Recession, Downturn has forced several Organizations to rethink there IT spending and make sure Business gets all the help they need. In my experience I have seen instances where CIO / CTO / Architects / IT Managers
- Spend majority of implementation time on business process engineering only to realize that the ERP solution which they purchased needs millions of dollars of customizations.
- Taking years building frameworks or standards which gets out dated everytime its used.
- Choosing products which are so expensive and which needs consultants to do any implementation
- Jump on to bleeding edge solutions where there is no need and no expertise. At one of my clients as part of SOA implementation the team ended up hooking up two systems with all sorts of web services only to realize the two applications are hosted in same data center next to each other.
- Start building a data center when there is a lack of infrastructure, unix / web administrators
These mis steps can be avoided if CIO/CTO & IT managers are tech savvy and look at alternatives are a snapshot of options listed below
- Make sure there are Open source alternatives for the business issues. There are already ERP/CRM, IT service desk systems which are comparable to product suites. SugarCRM, Compiere, ofBiz, OpenTaps and IT service desk systems such as OTRS are few
- Go in for a SAS model instead of building an infrastructure and having teams to manage your IT operations. For example Google, SalesForce.com, Zoho offer several applications which can help your Organization
- Buy against build. Try out products before you start building your own application or system and once again check out on Opensource
- Start Simple. Do not go with big bang approaches for your SOA, Content / Portal solutions. Try with in, use simple integration approaches such as Spring Integration, Mule, Content solutions such as Alfresco, Magnolia, LifeRay are few
- In House vs Out Source. Instead of building an application try to out source it, check out what free lancers can provide. elance is an excellent site for companies
In short this is what CEO wants from his IT leaders. The IT leadership which can think out of box and make sure business needs
Now with Cloud computing the latest IT buzz, Companies want to make sure they invest right and end up with returns. Though Cloud computing is making rounds Companies have to plan before they jump into that bandwagon. Some of the factors are
- Need for Cloud. Why are they opting for Cloud. Will it be better than a hosting solution? Does it help them Self Service or will it help them to charge others for pay by use.
- Choosing a Cloud. There are several cloud vendors but most of them serve their niche areas. For example Amazon ec2 is one of the best of Infrastructure Cloud. Similarly Azure for MS and its weave of products. CIO’s should be very clear on what they are looking for and what skills they have. For example Google cloud requires some basic knowledge of python
- Type of Cloud: Based on the business model companies need to understand whether they should opt for a public cloud or private cloud or a community cloud.
CIO’s should start understanding basic factors driving Cloud such as
- Scaling your apps or infrastructure
- Pay by use or subscription
- Self service (without system / db admins)
- programmability (easy access to build apps)
Based on the findings Companies can start with baby steps.
- Start with a due diligence study to understand the requirement
- Use PAAS or IAAS as a start
- Slowly move applications to a Private Cloud as need dictates.
- Understand security concerns and try the same on a public cloud.
Two good references to start on cloud computing are infoworld deep dive on cloud computing http://www.infoworld.com/d/cloud-computing/download-the-cloud-computing-deep-dive-report-500 and Demystifying the cloud by Janakiram MSV http://www.janakiramm.net/blog/ .
Managed Services are the next wave for Companies who are looking to wrap up their implementation. Most of the Companies who have standardized on their core products now are looking for managed services partner MSP who can provide stable support at a predefined cost. Signing up for a MSP provides a coverage guarantee of predefined services and KPI’s
For a provider who provides MSP the challenge is to provide excellent service at low cost. The provider has to be innovative and adopt repeatable process, best tools and practices to make sure they can provide TCO to their clients. Here are few tips to be a successful MSP
- Proper Services Agreement in place. The agreement should have Terms, payments, coverage hours, exclusions, standards, response and resolution, liabilities, confidentiality agreement.
- Have Problem Management, Incident Management systems in place. Leave no stones unturned to make sure your staff gets alerted for severity tickets.
- Major business flow needs to be displayed across the floor. L2 support can be resolved quickly if functional users are aware of the process ticket resolution will be quicker.
- Reward the team for building successful knowledge base. If team member adds a repeatable problem or solution reward him. This will make sure your knowledge base is robust enough to resolve majority of issues at L1 or help desk level.
- All Managed services have 4 major phases such as initiation, transition, stabilization and Improvement. As part of ROI, MSP should help clients in suggesting modernization, transformation, integration and/or performance as part of Improvement.
These basic steps will take a MSP to be a partner of a client. Follow this mantra for successful relationship
Google, Facebook and Amazon are leaders in their own area. Now each one wants to venture into new business or expand into areas. Google is trying to get into sweet spot of Facebook using several means such as enhancing orkut, google me, buzz etc.
What Should Google do now
Goolge is a leader with its maps, youtube and android. It should continue on path of innovation
- Improve the search capabilites
- Keep the market share slipping away to Bing
- Collate the reviews of the products such as google products
- Move forward with Google voice
- Make sure Chrome becomes an alternate OS
Challenge Facebook by acquistions such as
- Linkedin which connects professionals and has potential to grow
- Craigslist which is a network of online community
- Look at acquiring twitter
Unlike search, Social networking is connecting individuals and groups. Google should look at in interest areas where these individuals connect. Also look at buying out eBay which brings them into a auction business and stare at Amazon.
Google should be hardware agnostic and should stop becoming a hardware vendor such as rebranding HTC phones, Chrome laptops. Innovation should be the priority rather than managing helpdesk for these products. I invite all of reader community comments
Last few years I have seen Companies coming up with different delivery models and frameworks. Some have factories, competencies and COE supporting them.Few of these models have worked and some have remained in the Presentations to age along.
Lot of clients, captives still feel pressure of Project / Program Delivery due to issues such as
Poor Project Management
Lack of Technical / Architecture skills
One of the successful ways to address these issues is to have a shared model or a competency model. The Competency model helps in terms of scaling up and technical agility. Competency team can support multiple projects and technologies. Competency would help a project which needs expertise, provide best practices and also keeps project on schedule.
Setting up of a Competency does not come easy. It requires thought leadership with investments with focus on emerging technologies, R&D and constant training. Competency in itself will be a SWAT team helping Projects at right time. Comptency or COE is best explained by Madhusudan Rao my colleague