dr. shankar-dev ops-davinci_code_gtr

dEv ops: decoding the da Vinci code using PANCHATANTRA OF DEVOPS

BY DR SHANKAR RAMAMOORTHY Quality business partnerAmdocs testing services

2Information Security Level 2 – Sensitive© 2016 – Proprietary and Confidential Information of Amdocs

Da Vinci Code

• Da Vinci Code is a 2003 mystery detective novel by Dan Brown. It follows symbologist Robert Langdon and cryptologist Sophie Neveu after a murder in the Louvre Museum in Paris, when they become involved in a battle between the Priory of Sion and Opus Dei over the possibility of Jesus Christ having been married to Mary Mary Magdalene The title of the novel refers, among other things, to the finding of the first murder victim in the Grand Gallery of the Louvre, naked and posed like Leonardo da Vinci 's famous drawing, the Vitruvian Man with a cryptic message written beside his body and a pentagram drawn on his chest in his own blood.

• DevOps in this presentation is being seen as complicated as Da Vinci Code in this presentation. Cracking the code needs a story telling which has been attempted in the form of Panchatantra where fables of Panchatantra are compared to the journey of DeVOps.

• Creating a delivery system( Systems Thinking) , with business value and delivery pipeline, amplifying the feedback loops and learning and experimenting continuously using the pillars of Devops namely Culture, processes and technology starts with a common story line

• Customer requirements come at short intervals, developer codes requirements and commit changes, ongoing and automatic build in 2 hours, automatic deployments and testing on E2E prod like environment and deploying of package on UAT E2E environment with seamless deployment to production. Where is the mystery, I cracked this and found

“ Everything is automatic, Everything is versioned and Everything is Code AND that is the DaVinci Code for unlocking the DevOps approach.


Da Vinci Code • Gartner predicts that by 2017, 75% of large enterprise IT organizations will go BiModal and run on

two engines, engine of stability and fluidity.One will move fast and another will have to move slow.

• With DevOps everyone is a developer and everything is code and that is the Da Vinci Code

• DevOps is quickly becoming one of the most valuable disciplines for your business. It’s focused on improving the quality and speed of delivering new apps to market. And it’s about tightly integrating development and operations in order to do so. The average organization adopting DevOps sees a 20 percent improvement in time-to-market, a 22 percent improvement in software quality, and a 17 percent improvement in frequency of application deployments—all leading to 22 percent more customers and a19 percent increase in revenue

• Right from commiting a code and using P4 for version control, to Jenkins/Maestro for Orchestrator and CI server, to using Apache Maven and Gradle for Build and using of Junit/Mockito/PowerMock and Jasmine to test, to using Nexus for artifacts repository to SonarQube for Code metrics, Check Marx for Security to roaring Tiger for Database and Xpi for Deployment with a Vmware serving Environment As A Service, a whole set of CI ecosystem has challenged the old order and Continous Delivery Model ( CDM) challenging our GDM( Global Delivery Model )

• In an application economy, every business is software business and everything is code.


Panchatantra of DevOps• The Panchatantra is a series of inter-woven colourful tales, mostly fables. According to its

introductory narration, these inter-woven tales are meant to illustrate, for the benefit of three ignorant princes, the primary Hindu principles of nïti - "the wise conduct of life" or "prudent worldly

conduct". The three ignorant princes were DeV, QA, Ops.

• In a brief introductory narrative - Vishnu Sharma, is introduced as reciting and explaining the rest of the work to the three young princes. The book is then divided into five parts - each of which revolves around a main frame story, with several stories "emboxed" in it, sometimes three or four levels deep, as one character narrating a story to another. Often, these emboxed stories snap from each other, unexpectedly and irregularly at times, to sustain attention. Apart from the stories, the characters are noted to quote various epigrammatic verses to illustrate their objectives.

• The journey of DevOps can be traced similar to Panchatantra where the main actors as in Dev, QA and Ops can marry, divorce and then re marry per “ Niti” – Wise conduct of Life.

• Panchantra also explains that Niti and Sandarb ( Context) are conjoint twins and best explained by the author as “ a new race of Context Imperial Developers and Testers”


Panchatantra 1- Mitra Bheda

• The Separation of Friends (The Lion and the Bull) The first book, which is the longest of the five books, illustrates the incidents that frequently lead to separation of friends. It begins with a friendship between a lion king, Pingalaka, and a bull, Sanjivaka. Two jackals, Karataka and Damanaka, are ministers of the lion who lead to breaking up the friendship. The jackals narrate most of the thirty stories contained in this part. This book accounts for roughly 45% of the work's length.

• Corollory: Dev and QA were together 30 years ago in EDP Departments or MIS departments of various Government and Industrial Enterprises. Here too, the COBOL code was written, tested and productionized into vast array of Legacy Systems. The closed systems. IV&V practice, TCOE practice and Test Factory and Managed Test Services separated the friends.


Panchatantra 2 – Mitra Labha or Mitra Samprapti

• The Gaining of Friends (The Dove, Crow, Mouse, Tortoise and Deer)

• This second book illustrates the making or gaining of friends through the friendship between a rat and a crow, which later includes a turtle and a fawn. They collaborate to save each other to illustrate the essence of unity and friendship. This book accounts for roughly 22% of the total work's length.

• Corollory : Agile Practices , Agile Manifesto, Agile Culture made the warring factions colloborate to present to customer as a single entity interwoven by spirits of Quality and Excellence


Panchatantra 3: Kakolukiyam

• Of Crows and Owls (War and Peace) The third book illustrates the art of war and the importance of peace through traditional enemies - the crows and the owls. The book showcases the weaker crows overcoming the mightier owls by tactfully gaining access to their secrets and vulnerabilities. This book accounts for roughly 26% of the total work's length.

• Corollory: The testers learning scripts, Testers shifting left, Testers becoming pseudo developers.


Panchatantra 4: Labdhapranasam

• Loss of Gains (The Monkey and the Crocodile) The fourth book enlightens how gains achieved may be lost through a symbiotic friendship between a monkey and a crocodile, who conspires to acquire the heart of the monkey. However, the monkey gets to know of this and uses wisdom to avoid the grim fate.

• Corollory : Symbiotic friendship of Dev and QA can result in loosing of identity of Testers as a separate species and no one knows when the “ monkey” in a tester will come to know or when the wisdom prevails, we can see Loss of Gains taking place.


Panchatantra 5: Apariksitakarakam• ILL-Considered Actions (The Brahmani and the Mongoose)

• The fifth book revolves around the cases that are considered unacceptable through a Brahmani and her pet mongoose. The mongoose defends her child albeit injuries, but gets punished due to the Brahmani's misunderstanding. She later regrets having killed her friend, and her child's

saviour.


Panchatantra of Success using Dev OpsDevOps is not a product, or even a particular technology. DevOps is a methodology that unites the often separate functions of software development (Dev) and production and operations (Ops) into a single, integrated, and continuous process. DevOps is about breaking down the barriers between Dev and Ops. It leverages people, processes, and technology to stimulate collaboration and innovation across the entire software development and release process. Dev and Ops must act and feel like they are a single team. S.No Problem Statement Response

1One does not discover software defects until late in the lifecycle—or worse, in production. “Get Ops Involved Early”

2All use Agile to speed development, but any gains evaporate once the app goes into production.

“Automate the release of new code into production, which include documentation”

3Our developers and testers are constantly waiting to access the resources they need, causing delays.

- Dispel the mindset “Development’s Job Is Finished Once the App Is in Production”

4We always pinpoint problems across development, testing, and production operations. “Connect the three parties”

5One sees simple human errors wreaking havoc during development and deployment. “Carry out Service Virtualization”

Microsoft Word Document


The value of Adopting DevOps can be Significant


To Deploy DevOps in a two-Speed IT environment, company needs to pay

attention to


Case study @NetFlix-Running at two speeds

Netflix, for instance, has created a cloud-based IT architecture that allows its developers to launch hundreds of software changes a day. Its website comprises hundreds of micro services hosted in the cloud, and each service is maintained by a dedicated DevOps team. Developers don’t need to request resources from the IT operations team; instead they can automatically build pieces of code into deployable web images. As those images are updated with new features or services, they can be integrated with Netflix’s existing infrastructure using a custom-built, web-based platform on which infrastructure clusters are created. Testing is carefully done in the production environment with a subset of users. Once the web images are live, a load-balancing technology routes part of the traffic to them from older versions. Automated monitoring ensures that if something goes wrong with the deployment of new images, traffic is routed back to older versions, and the new images are rolled back. Because of this level of automation, Netflix can deploy new code into its production environment within hours, where most companies would need months.


Managing a two-speed IT architectureA two-speed IT architecture allows large-scale companies to accelerate the release of innovative products and applications that could make a substantial difference to customers while maintaining legacy IT systems that are less innovative but still necessary for the stability of the business. This sort of architecture emphasizes tight integration between the software applications being developed and the hardware infrastructure supporting them. Historically the IT operations teams maintaining software and hardware have been kept entirely separate. But with the increasing prevalence of vertical enterprise-resource-planning systems, the advent of network virtualization, and the emergence of software-as-a-service models, the two sides have moved closer to one another. These technology trends have had the effect of removing complexity from hardware stacks and making them more accessible to software developers.


Managing a two-speed IT architectureA two-speed environment requires that companies introduce automation tools to support continuous delivery of software—especially in the testing and production phases. Automation can allow for better management of, among other things, the release of software updates, the porting of new code, and the general processing environment. Most important, automation tools and cloud-based technologies can serve as the bridge between legacy IT systems on the back end and consumer-facing applications on the front end, allowing companies to pursue seamless testing, provisioning, deployment, governance, and security of servers and new software releases


IT is possible to deploy a new code on a site within an hour


IT is possible to deploy a new code on a site within an hour

A two-speed IT architecture conveys a number of critical advantages, but it takes time, careful consideration, and commitment to establish. Netflix, for instance, developed most of its cloud and automation technologies in-house, but companies have any number of products and packages (some open source) to choose from that can allow them to achieve similar dual-speed performance.The most critical factor in establishing a two-speed architecture is for IT leaders to adopt a capabilities-based view of the IT architecture, rather than a system- or process-oriented view. Instead, IT leaders could allocate resources toward “fast” and “slow” applications as required—gaining the critical benefits of the DevOps approach where it is possible to do so.


Managing a two-speed IT organization

The DevOps approach challenges the established product-development norms in most IT organizations. Additionally, IT leaders adopting DevOps organizational models may need to reconsider how technology partners are integrated into their software-delivery processes—a trend that is forcing some system vendors to consider ways to make their platforms available as a service.


Managing a two-speed IT organization

The biggest task for IT leaders is to identify those parts of the company where the use of DevOps would make most sense—likely focusing on those parts of the business where speed is at a premium, and where there is a significant opportunity for the company to differentiate its customer experience from the competition. (Think of a retailer using DevOps to improve its website checkout experiences, or a bank offering new fund-tracking capabilities at its site.) For those parts of the business where DevOps might make less sense—where reliability and resilience of software is more important than speed to market—IT leaders will need to determine how to maintain the split between software development and IT operations, and which roles and processes to adapt for a culture of continuous delivery.


Cross-functional teams of application-development, infrastructure-management, and operations professionals should be convened to streamline the ownership of stacks across the application-delivery pipeline. In the case of continuous delivery, for instance, a joint team would oversee all processes (and associated tools) relating to this development activity, such as application building, testing, and deployment; performance management and monitoring; and virtualization and configuration management.

Redefined roles- A generalizing Specialist


Redefined Culture and Talent

In a DevOps environment they will be held responsible for the quality of their code. They will need to know operating-system basics and must show strong collaboration skills as they work jointly with operations engineers to determine how best to solve application-development or deployment problems. As a result, many companies are already modifying their recruiting practices to hire “full stack” engineers—professionals who understand all aspects of computing, including user interfaces, databases, and networks.


Redefined processes and governance

Companies may want to look across the entire spectrum of software-delivery processes to determine which will need to be redefined or fully automated so that development teams can take advantage of infrastructure as a service, as needed, and so that code can be ported into testing and production environments in a standardized way.For instance, Internet companies enforce “self service” for developers; teams can test, promote, and deploy code in production environments without requiring constant hands-on involvement from infrastructure-operations teams, although both teams share responsibility for code performance.


Thank you• The value of implementing DevOps can

be significant with respect to both productivity and time to market. Knowing the Panchatantra and two speed engine is key for driving this vehicle to achieve Quality Goals and better Customer Experience and Loyalty drivers

dr. shankar-dev ops-davinci_code_gtr

Technology