Microservices Manchester: Microservices and Macro-Economics - A Shorty History By Marco Cullen
1 likeā¢630 views
The document outlines the historical evolution of microservices and their relationship with macroeconomics, emphasizing the distinction between large and small programming efforts. It discusses the integration of computer science and engineering, the importance of clear project parameters, and how microservices facilitate productivity through modular design. Additionally, it highlights the broader impact of microservices on business interactions, project management, and cost efficiency.
![IS THAT MY SIZE? IT FITS PERFECTLY
āø LOC output varies by 2 orders of magnitude
āø a 1000 lines / dev-weak
āø a 1000 lines / dev-year [Windows 95]
āø There is a wide range of productivity between projects
āø What inļ¬uences or contributes to this?
āø http://martinfowler.com/bliki/
CannotMeasureProductivity.html](https://image.slidesharecdn.com/microservicesandmacroeconomicsmarcocullen-160719100829/85/Microservices-Manchester-Microservices-and-Macro-Economics-A-Shorty-History-By-Marco-Cullen-29-320.jpg)
Microservices Manchester: Microservices and Macro-Economics - A Shorty History By Marco Cullen
- 1. MICRO SERVICES AND MACRO ECONOMICS - A SHORT HISTORY HOW DID WE GET HERE?
- 2. WHAT CAME BEFORE ..TO BUILD THE HISTORY OF STRUCTURE āø āthe quality of being organised"
- 3. WHAT CAME BEFORE ..TO BUILD THE HISTORY OF STRUCTURE āø āthe quality of being organised" āø Programming-in-the-large versus programming-in-the- small (Hans Kron and Frank DeRemer 1975)
- 4. WHAT CAME BEFORE ..TO BUILD THE HISTORY OF STRUCTURE āø āthe quality of being organised" āø Programming-in-the-large versus programming-in-the- small (Hans Kron and Frank DeRemer 1975) āø We distinguish the activity of writing large programs from that of writing small ones. By large programs we mean systems consisting of many small programs (āmodulesā) written by different people, possibly in different languages.
- 5. WHAT CAME BEFORE ..TO BUILD THE HISTORY OF STRUCTURE āø āthe quality of being organised" āø Programming-in-the-large versus programming-in-the- small (Hans Kron and Frank DeRemer 1975) āø We distinguish the activity of writing large programs from that of writing small ones. By large programs we mean systems consisting of many small programs (āmodulesā) written by different people, possibly in different languages. āø Module Interconnection language
- 6. 2016 MODULES!1976 MICRO SERVICES! WHAT HAVE WE LEARNT IN 40 YEARS?
- 7. COMPUTER ENGINEERING VS COMPUTER SCIENCE āø Deļ¬nitions āø Computer science is the study of the principles and use of computers.
- 8. COMPUTER ENGINEERING VS COMPUTER SCIENCE āø Deļ¬nitions āø Computer science is the study of the principles and use of computers. āø Computer engineering is a discipline that integrates several ļ¬elds of electrical engineering and computer science required to develop computer hardware and software.
- 9. COMPUTER ENGINEERING VS COMPUTER SCIENCE āø Deļ¬nitions āø Computer science is the study of the principles and use of computers. āø Computer engineering is a discipline that integrates several ļ¬elds of electrical engineering and computer science required to develop computer hardware and software. āø Module Interaction Language
- 10. COMPUTER ENGINEERING VS COMPUTER SCIENCE āø Deļ¬nitions āø Computer science is the study of the principles and use of computers. āø Computer engineering is a discipline that integrates several ļ¬elds of electrical engineering and computer science required to develop computer hardware and software. āø Module Interaction Language āø Representational State Transfer application programming interface, RESTful API
- 11. THE NATURAL WAY OF THINGS - REWINDING THE TAPE OF SOFTWARE āø Copes rule āø āā¦.over the past 542 million years, the mean size of marine animals has increased 150-fold.ā Jonathan Payne, a paleobiologist at Stanford School of Earth, Energy & Environmental Sciences
- 13. THE NATURAL WAY OF THINGS - REWINDING THE TAPE OF SOFTWARE āø Copes rule āø āā¦.over the past 542 million years, the mean size of marine animals has increased 150-fold.ā āø āFor reasons that we donāt completely understand, the classes with large body size appear to be the ones that over time have become differentially more diverse.ā Jonathan Payne, a paleobiologist at Stanford School of Earth, Energy & Environmental Sciences
- 14. THE NATURAL WAY OF THINGS - REWINDING THE TAPE OF SOFTWARE āø Copes rule āø āā¦.over the past 542 million years, the mean size of marine animals has increased 150-fold.ā āø āFor reasons that we donāt completely understand, the classes with large body size appear to be the ones that over time have become differentially more diverse.ā āø āItās really a story of the survival and diversiļ¬cation of big things relative to small thingsā Jonathan Payne, a paleobiologist at Stanford School of Earth, Energy & Environmental Sciences
- 15. ENSURING SUCCESS FROM HUTS TO HOUSES
- 16. A LORD A FOREMAN AN ARTISAN 3 PROPERTIES THAT SUCCESSFUL PROJECTS OFTEN SHARE āø Person of authority āø Process āø Technical authorship through expertise and experience Harvard Business School - āā¦Factors common to successful projectsā
- 17. A LORD A FOREMAN AN ARTISAN 3 PROPERTIES THAT SUCCESSFUL PROJECTS OFTEN SHARE āø ā¦. a project will not fail because of the type of tools used or the type of methodology employed Harvard Business School - āā¦Factors common to successful projectsā
- 18. A LORD A FOREMAN AN ARTISAN 3 PROPERTIES THAT SUCCESSFUL PROJECTS OFTEN SHARE Citation Frequency Clear requirements and speciļ¬cations 60.5 Clear objectives and goals 55.8 Realistic schedule 53.5 Effective project management skills - PM 53.5 Support from top management 51.2 User/client involvement and feedback 46.5 Realistic budget 46.5 Skilled and sufļ¬cient staff 44.2 Frozen requirement 41.9 Critical success factors for software projects: A comparative study
- 19. WHY CANT WE MAKE SOFTWARE LIKE WE MAKE HOUSES?
- 21. EVOLUTIONARY ARCHITECTURE ITS MORE THAN JUST A HOUSE
- 22. INSPIRATION FROM NATURE āø Cellular āø Iterative āø Specialised āø Bounded āø Obligated DOMAIN DRIVEN DESIGN CONTRACT DRIVENDESIGN BOUNDED FUNCTIONAL CONTEXT
- 23. AND SCALE MICRO SERVICES, DESIGN
- 24. DESIGN DESIGN DESIGN.. HOW DOES MICRO SERVICES FIT IN? āø Service can be deļ¬ned as āø domain bounded āø contract bounded
- 25. DESIGN DESIGN DESIGN.. HOW DOES MICRO SERVICES FIT IN? āø Service can be deļ¬ned as āø domain bounded āø contract bounded āø This has desirable scaling properties relative toā¦
- 26. HOW LONG IS A PIECE OF SPRINT
- 28. WHATāS A REAL MAN DAY?
- 29. IS THAT MY SIZE? IT FITS PERFECTLY āø LOC output varies by 2 orders of magnitude āø a 1000 lines / dev-weak āø a 1000 lines / dev-year [Windows 95] āø There is a wide range of productivity between projects āø What inļ¬uences or contributes to this? āø http://martinfowler.com/bliki/ CannotMeasureProductivity.html
- 30. IS THAT MY SIZE? IT FITS PERFECTLY āø Why is productivity hard to measure
- 31. IS THAT MY SIZE? IT FITS PERFECTLY āø Why is productivity hard to measure āø There is no deļ¬nition that captures holistically the value in a way that is both relevant to the client and to the business.
- 32. IS THAT MY SIZE? IT FITS PERFECTLY āø Why is productivity hard to measure āø There is no deļ¬nition that captures holistically the value in a way that is both relevant to the client and to the business. āø People agree functionality points are quite good but they donāt relate the value of what was delivered to the business with a view to future work, i.e. was their duplication, can we re-use?
- 33. IS THAT MY SIZE? IT FITS PERFECTLY āø Why is productivity hard to measure āø There is no deļ¬nition that captures holistically the value in a way that is both relevant to the client and to the business. āø People agree functionality points are quite good but they donāt relate the value of what was delivered to the business with a view to future work, i.e. was their duplication, can we re-use? āø Lets focus on process and maybe we get productivity?
- 34. INSPIRATION FROM NATURE āø Cellular āø Iterative āø Specialised āø Bounded āø Obligated DOMAIN DRIVEN DESIGN CONTRACT DRIVENDESIGN BOUNDED FUNCTIONAL CONTEXT
- 35. WHO KNOWS WHAT āø I know what i know, but not what he knows āø Heterogeneous structures require specialisation āø Different components evolve at different speeds āø How do we on board and off board specialist knowledge? āø Do i have to know what he knows to do what i know?
- 36. INSPIRATION FROM NATURE āø Cellular āø Iterative āø Specialised āø Bounded āø Obligated DOMAIN DRIVEN DESIGN CONTRACT DRIVENDESIGN BOUNDED FUNCTIONAL CONTEXT
- 37. A MACRO IMPACT MICRO SERVICES
- 38. WHATS AFFECTED āø Business interaction modalities āø daily routines - scrum, standup āø time and length of interactions āø Processes and benchmarks āø New processes to deliver functioning software at a high rate of mutation - Dev/Ops āø New automated methods for ensuring adherence to function - Acc. Tests āø Integration - CI/CD āø New metrics - performance testing* āā- SLA
- 39. A MORE NATURAL WAY TO THINK ORGANISE AND DELIVER MICRO SERVICES
- 40. WHAT IS AFFECTED āø ..running a micro services project results in various outcomes which implicitly help project management and thus the path to delivery āø Repeatable processes and thus a repeatable product
- 41. WHAT IS AFFECTED āø ..running a micro services project results in various outcomes which implicitly help project management and thus the path to delivery āø Repeatable processes and thus a repeatable product āø Well deļ¬ned areas of responsibility, knowledge and accountability
- 42. WHAT IS AFFECTED āø ..running a micro services project results in various outcomes which implicitly help project management and thus the path to delivery āø Repeatable processes and thus a repeatable product āø Well deļ¬ned areas of responsibility, knowledge and accountability āø A clear set of interaction modalities between all parties including the software and people alike!
- 43. WHATS NOT AFFECTED āø ā¦ so If my project is a micro services project, it WILL succeed? āø methodology and tooling not highly correlated with success.
- 44. A LORD A FOREMAN AN ARTISAN 3 PROPERTIES THAT SUCCESSFUL PROJECTS OFTEN SHARE Citation Frequency Clear requirements and speciļ¬cations 60.5 Clear objectives and goals 55.8 Realistic schedule 53.5 Effective project management skills - PM 53.5 Support from top management 51.2 User/client involvement and feedback 46.5 Realistic budget 46.5 Skilled and sufļ¬cient staff 44.2 Frozen requirement 41.9 Critical success factors for software projects: A comparative study
- 45. WHATS NOT AFFECTED āø ā¦ so If my project is a micro services project, it WILL succeed? āø methodology and tooling not highly correlated with success. āø ā¦ā¦they are however strongly correlated with cost*
- 46. WHATS NOT AFFECTED āø ā¦ so If my project is a micro services project, it WILL succeed? āø methodology and tooling not highly correlated with success. āø ā¦ā¦they are however strongly correlated with cost* āø ā¦but my operational overheads are higher now and i have duplication of effort, how do i end up saving on cost?
- 47. WHATS IS AFFECTED - THE MACRO OF MICRO āø Where are my costs reduced āø Reduction in cost to requirements change āø changes are often bounded, reducing complexity āø cross cutting changes are more easily decomposed āø Late changes cost less - a major contributor to cost and one of the top 10 reasons related to project success.
- 48. WHATS IS AFFECTED āø Where are my costs reduced āø Organisationally now we can deal with a high rate of mutation āø Able to change direction with lesser cost impact relative to more traditional monolithic structures. āø Synergises at scale with Agile, containerisation and established developer working and hiring patterns. WHATS IS AFFECTED - THE MACRO OF MICRO
- 49. WHATS IS AFFECTED āø Where are my costs reduced āø Working software is one of the best ways to measure progress. WHATS IS AFFECTED - THE MACRO OF MICRO School of Computer Science and software Engineering - The University of Western Australia 2012
- 50. WHATS IS AFFECTED āø Where are my costs reduced āø Working software is one of the best ways to measure progress āø Working software vs effort is 1 measure of productivity WHATS IS AFFECTED - THE MACRO OF MICRO School of Computer Science and software Engineering - The University of Western Australia 2012
- 51. WHATS IS AFFECTED āø Where are my costs reduced āø Working software is one of the best ways to measure progress āø Working software vs effort is 1 measure of productivity āø Micro Services results in working code deployed sooner than more traditionally linear approaches to product development. WHATS IS AFFECTED - THE MACRO OF MICRO School of Computer Science and software Engineering - The University of Western Australia 2012
- 52. A MACRO IMPACT ON YOUR BUSINESS ECONOMICS MICRO SERVICES