How did we end up here? Todd Montgomery Martin Thompson How bad - - PowerPoint PPT Presentation

▶
how did we end up here
SMART_READER_LITE
LIVE PREVIEW

How did we end up here? Todd Montgomery Martin Thompson How bad - - PowerPoint PPT Presentation

Sep 21, 2022 352 likes •1.44k views

How did we end up here? Todd Montgomery Martin Thompson How bad can things really be? Software Project Success Rates Successful Challenged Failure Ad-hoc 49% 37% 14% Iterative 61% 28% 11% Agile 60% 28% 12% Traditional 47% 36%

slide-1
SLIDE 1

How did we end up here?

Todd Montgomery Martin Thompson

slide-2
SLIDE 2
slide-3
SLIDE 3
slide-4
SLIDE 4
slide-5
SLIDE 5
slide-6
SLIDE 6

How bad can things really be?

slide-7
SLIDE 7

Successful Challenged Failure

Ad-hoc 49% 37% 14% Iterative 61% 28% 11% Agile 60% 28% 12% Traditional 47% 36% 17% Software Project Success Rates

  • Dr Dobbs 2010
slide-8
SLIDE 8

< 10 11 - 25 > 25

Ad-hoc 70% 58% 40% Iterative 88% 68% 55% Agile 83% 70% 55% Traditional 69% 51% 50% Software Project Success Rates – by Team Size

  • Dr Dobbs 2010
slide-9
SLIDE 9

Well that’s the

  • ptimistic view!
slide-10
SLIDE 10

Successful: 32% Challenged: 44% Failure: 24%

– Standish Group Chaos Report 2010

Software Project Success Rates

slide-11
SLIDE 11

< $350,000: 20% $350,000 - $1,000,000: 25% > $1,000,000: 28%

– Gartner 2012

Software Project Failure Rates

slide-12
SLIDE 12

In a study of over 5400 large scale projects (> $15m) 17% go so badly that they threaten the existence of the company undertaking them

– The McInsey Group with Oxford University 2012

slide-13
SLIDE 13

Sacred Cows - It’s BBQ Time!!!

slide-14
SLIDE 14

Enterprise Software

slide-15
SLIDE 15

ent•ter•prise

noun \ˈen-tə(r)-ˌprīz\ : a project or activity that involves many people and that is often difficult : the ability or desire to do dangerous or difficult things or to solve problems in new ways

Source: http://www.merriam-webster.com/

slide-16
SLIDE 16

Naming Matters !!!

slide-17
SLIDE 17
slide-18
SLIDE 18
slide-19
SLIDE 19

Product Management

slide-20
SLIDE 20

Minimum Viable Product ?

slide-21
SLIDE 21

Product Owner ?

slide-22
SLIDE 22
slide-23
SLIDE 23

Technologists ARE part of the business

slide-24
SLIDE 24

Take responsibility for ROI

slide-25
SLIDE 25

How can I get an answer for the minimum investment?

slide-26
SLIDE 26

Agile Methods

slide-27
SLIDE 27
slide-28
SLIDE 28
slide-29
SLIDE 29

Water-scrum-fall

slide-30
SLIDE 30

What really matters?

slide-31
SLIDE 31

Need to focus on learning, feedback cycles, and outcomes

slide-32
SLIDE 32

There is an uncomfortable truth…

slide-33
SLIDE 33

“What would have been different if you were not involved?”

slide-34
SLIDE 34

Manifestos

slide-35
SLIDE 35

Agile Reactive NoTCP

slide-36
SLIDE 36
slide-37
SLIDE 37

Build on the shoulders of giants

slide-38
SLIDE 38

Shared Mutable State

slide-39
SLIDE 39

“…Shared Mutable State...” the most feared words in computing

slide-40
SLIDE 40

…if not they should be!

slide-41
SLIDE 41
slide-42
SLIDE 42
slide-43
SLIDE 43

Shared Mutable State should only be used for systems programming

slide-44
SLIDE 44

Embrace append-only, single writer, and shared nothing designs

slide-45
SLIDE 45

If you don’t... math will hunt you down and there is nowhere to hide!

slide-46
SLIDE 46

Universal Scalability Law

2 4 6 8 10 12 14 16 18 20 1 2 4 8 16 32 64 128 256 512 1024

Speedup Processors

Amdahl USL

slide-47
SLIDE 47

Be ruthless in reducing complexity

slide-48
SLIDE 48

Text Encoding

slide-49
SLIDE 49
slide-50
SLIDE 50

But it’s human readable...

slide-51
SLIDE 51

Binary is hard to work with...

slide-52
SLIDE 52
slide-53
SLIDE 53
slide-54
SLIDE 54

Big Data

slide-55
SLIDE 55

Communications Battery life and bandwidth?

slide-56
SLIDE 56

Synchronous Comms

slide-57
SLIDE 57

Bad things will happen!!!

slide-58
SLIDE 58

Synchronous Communication is the crystal meth of distributed programming

slide-59
SLIDE 59

Causes a coupling in location and time

slide-60
SLIDE 60

Errors need to be first class messages

slide-61
SLIDE 61

Are your micro services

  • n crystal meth?
slide-62
SLIDE 62

Abstraction

slide-63
SLIDE 63

“All non-trivial abstractions, to some extent, are leaky.”

  • Joel Spolsky
slide-64
SLIDE 64

“The detail of underlying complexity cannot be ignored.”

slide-65
SLIDE 65

“the purpose of abstracting is not to be vague, but to create a new semantic level in which

  • ne can be absolutely precise”
  • Dijkstra
slide-66
SLIDE 66

We could say the main issue is that people don’t understand abstractions...but...

slide-67
SLIDE 67

Sins committed in the name of Abstraction

slide-68
SLIDE 68

ORMs !!!

slide-69
SLIDE 69

Functional Programming

slide-70
SLIDE 70

What is the biggest issue with functional programming?

slide-71
SLIDE 71

Functional Programmers

slide-72
SLIDE 72

Functional programming is

NOT

the answer to multi-core

slide-73
SLIDE 73

Software Transactional Memory was a failed experiment!

slide-74
SLIDE 74

Universal Scalability Law

2 4 6 8 10 12 14 16 18 20 1 2 4 8 16 32 64 128 256 512 1024

Speedup Processors

Amdahl USL

slide-75
SLIDE 75

No Mechanical Sympathy?

slide-76
SLIDE 76

However there is genuine brilliance in functional programming

slide-77
SLIDE 77

Collaborate and great things can happen...

slide-78
SLIDE 78

Throw hardware at it… development is too expensive

slide-79
SLIDE 79

The free lunch is over… we cannot be sloppy anymore…

slide-80
SLIDE 80

L0 1534 µops LB 28 µops

Loops

LB 28 µops

slide-81
SLIDE 81

Code must be simple and composable

slide-82
SLIDE 82

Cache Sub-System

L1(D) - 32K L2 - 256K L3 – 8-20MB LF/WC Buffers L1(I) – 32K L0(I) – 1.5k µops

32 Bytes / cycle

System Agent TLB Pre-fetchers TLB Pre-fetchers

128 bits / cycle

MOB Memory Controller QPI

QPI Bus Memory Channels 16 Bytes / cycle Ring Bus

PCI-e Controller

32 Bytes / cycle 32 Bytes / cycle

slide-83
SLIDE 83

Patterns of access and locality are key to performance

slide-84
SLIDE 84

Accumulated Improvement Time Bandwidth Latency

Memory Sub-System Performance

slide-85
SLIDE 85

Accumulated Improvement Time Network Bandwidth Response Time Storage Capacity CPU Cores Memory Capacity

slide-86
SLIDE 86

What does this mean for software?

slide-87
SLIDE 87
slide-88
SLIDE 88

Think in terms of transformation and flow of data – not code!

slide-89
SLIDE 89

Diversity

slide-90
SLIDE 90

Testosterone Driven Development

slide-91
SLIDE 91
slide-92
SLIDE 92

What did the Carnegie Mellon studies show?

slide-93
SLIDE 93
slide-94
SLIDE 94
slide-95
SLIDE 95

Fake it until you make it…

slide-96
SLIDE 96

“As soon as you realise that most people don’t know what they are doing the world makes a lot more sense…”

– Farley’s second law

slide-97
SLIDE 97

We need to look seriously at training programmers

slide-98
SLIDE 98

Coaching and Apprenticeships

slide-99
SLIDE 99

“The most important thing I've accomplished, other than building the compiler, is training young people.”

  • Grace Hopper
slide-100
SLIDE 100

“'Do you think we can do this?' I say, ‘Try it.’ And I back 'em up. They need that. I keep track of them as they get older and I stir 'em up at intervals so they don't forget to take chances.

  • Grace Hopper
slide-101
SLIDE 101

In closing…

slide-102
SLIDE 102

What are the greatest achievements

  • f the human race?
slide-103
SLIDE 103

The Scientific Method

slide-104
SLIDE 104

Understanding of Evolution

slide-105
SLIDE 105

Don’t feel bad… We are living in the era of Software Alchemy

slide-106
SLIDE 106
slide-107
SLIDE 107

Questions?

@mjpt777 – Martin Thompson @toddlmontgomery – Todd Montgomery

“It does not matter how intelligent you are, if you guess and that guess cannot be backed up by experimental evidence – then it is still a guess.”

  • Richard Feynman