Keynote Devops Days Amsterdam - Hacking IT, Culture over Code Bringing Devops into your Organization

Editor's Notes

1. http://mitpress.mit.edu/books/new-hackers-dictionary
2. DevOps is a cutlural cahnge that strives to improve the delivery of
3. CIO Journal printed by the WSJ - DevOps Is Great for Startups, but for Enterprises It Won’t Work—Yet http://blogs.wsj.com/cio/2014/05/13/devops-is-great-for-startups-but-for-enterprises-it-wont-work-yet/
4. Wall Street Journal – Enterprise DevOps Adoption Isn’t Mandatory — but Neither Is Survival http://blogs.wsj.com/cio/2014/05/22/enterprise-devops-adoption-isnt-mandatory-but-neither-is-survival/
5. I worked for MindSpring and Award winning, PC World pick, Top 5 C|Net, PC Magazine. Alex Trubek was our spokesperson We did a good job making users happy.
6. Delivered Internet Access (Dial-Up, ISDN, Cable, DSL) Web access web sites Email
7. I came to MindSpring through a
8. MindSpring Core Values & Beliefs / 14 Deadly Sins http://en.wikipedia.org/wiki/MindSpring
9. Customer Acquisition Everyone was competing for internet customers, tons of free trials, little customer loyatly Rapid Commoditization The Internet was becoming a household requirement and it was much like phone service as the offerings were the same with little differentiation Thin Margins In many cases it was more expensive to deliver access then what customers paid Demand was Enormous S Images Courtesy of Moniker Hill on Flickr - https://www.flickr.com/photos/monkerino/191842293/in/photolist-hXf2K-85Q2yw-95tW1z-4i1sqF-4i1stH-4i5xQb-4i5y3N-ank3ZY-egB3hj-4XVoiM-9R1Yz6-8Hyg43-34wBtZ-8KTxdF-xBpQG-4i1srn-9AXSQ-5NUTy9-94PV5S-8iAJHT
10. Internet Transit Prices (1998-2014) U.S. Internet Region http://drpeering.net/white-papers/Internet-Transit-Pricing-Historical-And-Projected.php Cost for Internet transit in the U.s was approximately $1200 per megabit per second in 1998 Today cost for internet transit in the US is about $.94 per Mbps
11. http://www.nist.gov/baldrige/ In the early and mid-1980s, many U.S. industry and government leaders saw that a renewed emphasis on quality was necessary for doing business in an ever-expanding and more competitive world market. But many U.S. businesses either did not believe quality mattered for them or did not know where to begin. The Malcolm Baldrige National Quality Improvement Act of 1987, signed into law on August 20, 1987, was developed through the actions of the National Productivity Advisory Committee, chaired by Jack Grayson. The nonprofit research organization APQC, founded by Grayson, organized the first White House Conference on Productivity, spearheading the creation of the Malcolm Baldrige National Quality Award in 1987. The Baldrige Award was envisioned as a standard of excellence that would help U.S. organizations achieve world-class quality. In the late summer and fall of 1987, Dr. Curt Reimann, the first director of the Malcolm Baldrige National Quality Program, and his staff at the National Institute of Standards and Technology (NIST) developed an award implementation framework, including an evaluation scheme, and advanced proposals for what is now the Baldrige Award. In its first three years, the Baldrige Award was jointly administered by APQC and the American Society for Quality, which continues to assist in administering the award program under contract to NIST. The Baldrige Criteria for Performance Excellence serve two main purposes: (1) to help organizations assess their improvement efforts, diagnose their overall performance management system, and identify their strengths and opportunities for improvement and (2) to identify Baldrige Award recipients that will serve as role models for other organizations. In addition, the Criteria help strengthen U.S. competitiveness by improving organizational performance practices, capabilities, and results facilitating communication and sharing of information on best practices among U.S. organizations of all types serving as a tool for understanding and managing performance and for guiding planning and opportunities for learning The Baldrige Criteria for Performance Excellence provide organizations with an integrated approach to performance management that results in delivery of ever-improving value to customers and stakeholders, contributing to organizational sustainability improved organizational effectiveness and capabilities organizational and personal learning
12. Quality of Service - Connection speeds Speed to Response – How Quickly we answered the phone – If we answered a phone call in under 45 seconds customer satisfaction was very high, 45 seconds to 10 minutes on hold and satisfaction was relatively the same First Call Resolution was not as big a factor in customer satisfaction but provided a huge reduction in waste Self Service was also a big factor, change password, update account information online
13. http://gapingvoid.com/corporate-culture/
14. Photo courtesy of Matt Moor on Flickr http://www.flickr.com/photos/mattmflickr/7453204160/in/photostream/
15. http://www.inc.com/encyclopedia/corporate-culture.html
16. Most people say what we do and how we do it. Most tangible to the Apple Computer wants to challenge the status quo, they think different. We believe that in everything we do we should challenge the status quo. The way we we challenge the status quo is through simple, elegant design. We just happen to make great electronics What is the why for your organization? We do what we do so people can accomplish awesome stuff through the use of technology How we do that is by providing good coordination between developer and operations so that We provide automated, scalable, fault tolerant solutions to people. People don’t invest in what you do, why you do it.
17. What is the Why of your Organization? Deliver services so that you can do awesome stuff. At Apache CloudStack I like to think about our Cloud users as people who are proving the existing of the Higgs Boson or running culture How Open source devlopment model to make our IT more efficient What We create cloud computing software to enable users to access the resources they need on demand
18. Courtesy Doc Searls - https://www.flickr.com/photos/docsearls/5500714140/in/photolist-dSA9Bi-auR8jn-fu5X2f-bTHTce-c2m9DU-5GMvAV-E99M-5XDYbN-9o5AEY-cPkrKh-E9aS-fpKoDq-hq1y1M-4dYuXC-27hTDj-fyssX2-bVVC9G-E9c6-7FwX1A-kTNGSK/ via https://creativecommons.org/licenses/by/2.0/
19. Overproduction http://www.pdx.edu/fadm/sites/www.pdx.edu.fadm/files/02_Introduction%20to%20Lean%20Principles%20-%20Supergraphic.pdf http://www.leanproduction.com/intro-to-lean.html http://www.leanproduction.com/intro-to-lean.html Making something before it is truly needed. This is a particularly serious form of waste because it leads to excess inventory that is often used to mask other underlying problems and inefficiencies. Pace production so the rate of manufacturing matches the rate of customer demand (Takt Time). Use a pull system to control how much is manufactured (Kanban). Reduce setup times so that smaller batches can be economically manufactured (SMED). Waiting Time when work-in-process is waiting for the next step in production (no value is being added). It can be truly illuminating to look at the time from order to shipment and ask – how much of that time is actually spent on true value-added manufacturing. Design processes so that the flow is continuous and there are minimal (or no) buffers between steps in production (Continuous Flow). Use standardized work instructions to ensure that a consistent method and consistent times are used for each step of production (Standardized Work). Transport Unnecessary movement of raw materials, work-in-process or finished goods. Design a linear, sequential flow from raw materials to finished goods (Value Stream Mapping). Make sure work-in-process is not placed into inventory (Continuous Flow). Avoid continual changing of job priorities (Theory of Constraints). Motion Unnecessary movement of people (movement that does not add value). Ensure that work areas are logically organized (5S). Consider alternate arrangements of equipment that reduce motion (Value Stream Mapping). Overprocessing More processing than is needed to produce what the customer requires. This is often one of the more difficult wastes to detect and eliminate. Compare customer requirements to manufacturing specifications (Kaizen). Look for potential simplifications to the manufacturing process (Kaizen). Inventory Product (raw materials, work-in-process, or finished goods) quantities that go beyond supporting the immediate need. Bring raw materials in only as they are needed (Just-In-Time). Reduce or eliminate buffers between steps in production (Continuous Flow). Refer to Overproduction countermeasures (Takt Time, Kanban, and SMED). Defects Production that is scrap or requires rework. Design processes so they are less likely to produce defects (Poka-Yoke). Design processes to detect abnormalities so they can be immediately corrected (Jidoka). Look for the single most frequent defect and determine why it occurs (Root Cause Analysis). Create work instructions that provide a consistent method of manufacturing the part. (Standardized Work). Lean concepts become a lot more intuitive and easy-to-understand when they are traced to the ultimate goal – eliminating waste. verproduction Making something before it is truly needed. This is a particularly serious form of waste because it leads to excess inventory that is often used to mask other underlying problems and inefficiencies. Pace production so the rate of manufacturing matches the rate of customer demand (Takt Time). Use a pull system to control how much is manufactured (Kanban). Reduce setup times so that smaller batches can be economically manufactured (SMED). Waiting Time when work-in-process is waiting for the next step in production (no value is being added). It can be truly illuminating to look at the time from order to shipment and ask – how much of that time is actually spent on true value-added manufacturing. Design processes so that the flow is continuous and there are minimal (or no) buffers between steps in production (Continuous Flow). Use standardized work instructions to ensure that a consistent method and consistent times are used for each step of production (Standardized Work). Transport Unnecessary movement of raw materials, work-in-process or finished goods. Design a linear, sequential flow from raw materials to finished goods (Value Stream Mapping). Make sure work-in-process is not placed into inventory (Continuous Flow). Avoid continual changing of job priorities (Theory of Constraints). Motion Unnecessary movement of people (movement that does not add value). Ensure that work areas are logically organized (5S). Consider alternate arrangements of equipment that reduce motion (Value Stream Mapping). Overprocessing More processing than is needed to produce what the customer requires. This is often one of the more difficult wastes to detect and eliminate. Compare customer requirements to manufacturing specifications (Kaizen). Look for potential simplifications to the manufacturing process (Kaizen). Inventory Product (raw materials, work-in-process, or finished goods) quantities that go beyond supporting the immediate need. Bring raw materials in only as they are needed (Just-In-Time). Reduce or eliminate buffers between steps in production (Continuous Flow). Refer to Overproduction countermeasures (Takt Time, Kanban, and SMED). Defects Production that is scrap or requires rework. Design processes so they are less likely to produce defects (Poka-Yoke). Design processes to detect abnormalities so they can be immediately corrected (Jidoka). Look for the single most frequent defect and determine why it occurs (Root Cause Analysis). Create work instructions that provide a consistent method of manufacturing the part. (Standardized Work). Lean concepts become a lot more intuitive and easy-to-understand when they are traced to the ultimate goal – eliminating waste.
20. We have a dirth of automation tools at our disposal and many o Ansible Foreman - http://theforeman.org/ - Foreman is a complete lifecycle management tool for physical and virtual servers. Func -
21. Don’t just automate production, automate test, automate recovery Chaos Monkey http://bob.ippoli.to/erl_testing_2011/ http://appirio.com/category/tech-blog/2013/03/deliver-value-with-unit-testing-in-the-cloud/
22. CIO.gov - https://cio.gov/performance-metrics-and-measures/ There is overlap between measures and metrics. Both can be qualitative or quantitative, but what distinguishes them is important. Measures are concrete, usually measure one thing, and are quantitative in nature (e.g. I have five apples). Metrics describe a quality and require a measurement baseline (I have five more apples than I did yesterday). In a Section 508 program, measures are useful for demonstrating workloads and activity, and metrics are useful for evaluating compliance, processes effectiveness, and measuring success against established objectives.
23. Cloud computing promises highly available systems, but if you have a reactive approach you won’t achieve that goal. If you want a five nines service level you have 5.26 minutes to find, fix and recover Build redundant, highly environment systems
24. Visualizations of Continuous Delivery - http://continuousdelivery.com/2014/02/visualizations-of-continuous-delivery/ Nhan Ngo, a QA engineer at Spotify, made four fabulous visualizations while reading Continuous Delivery. She has very kindly agreed to make them available under a Creative Commons license so feel free to share them, download them, and print them out (click to get a higher resolution version). - https://www.linkedin.com/pub/nhan-ngo/11/938/ba5