10 Traumas in 10 minutes

A lightening talk - walking through 10 trauma's for developing applications for the cloud

Transcript

1. Hi
   

   View Slide

2. @ammeep
   

   View Slide

3. View Slide

4. 10
   E Traumas
   in 10 minutes
   

   View Slide

5. 1
   E
   Deployments
   are painful
   

   View Slide

6. automated
   deployments are hard in the cloud
   

   View Slide

7. automated
   deployments are hard in the cloud
   dirty lies
   

   View Slide

8. automated build
   automated test
   automated deploy
   

   View Slide

9. automate
   provisioning
   API
   

   View Slide

10. automate
    network setup
    API
    

    View Slide

11. automate
    deployment
    API
    

    View Slide

12. automate
    a the thing
    

    View Slide

13. 2
    E
    i’m married
    to my cloud
    

    View Slide

14. AbstractProxyBooleanProvider
    avoid a ma iage of inconvenience
    just don’t commit
    be abstract
    

    View Slide

15. abstract message queues
    abstract
    storage providers
    abstract
    auto scaling
    

    View Slide

16. abstract away
    the gory details
    if you don’t own it
    

    View Slide

17. 3
    E
    insanity
    the definition of
    

    View Slide

18. Oh HAI
    please may i retrieve
    my data?
    sure ... i’ll hold...
    

    View Slide

19. hello?
    ... ok holding ...
    may i have my data?
    

    View Slide

20. data?
    what?
    ........
    

    View Slide

21. View Slide

22. Are we there yet? Are we there yet? Are we there yet? Are we there yet?
    Are we there yet? Are we there yet? Are we there yet? Are we there yet?
    not a
    are born equal
    transient faults
    

    View Slide

23. exponential
    the nature of the transient fault
    Unde tand
    perhaps try
    back off
    

    View Slide

24. 4
    E
    storage
    the limitations
    of
    

    View Slide

25. Transactions can be capped
    but can we still have
    infinite storage capacity
    infinite throughput
    and
    

    View Slide

26. many ways to skin a cat
    When is it accessed
    Wh is accessing it
    What is accessed
    

    View Slide

27. 111001001100101011100110110010101101110011101000110000101110100011010010110111101
    011111100100000010000010110110100100000011010010010000001110100011010000110000101
    0010000001100010011011110111001001101001011011100110011100111111
    110011001110101011000110110101101101001011011100110011100100000011000110111010101
    111010000100000011000100110000101101100011011000111001100100000011110010110111101
    010000001100011011000010110111000100000011100100110010101100001011001000010000001
    110100001101001011100110011111100100000011101110110100001111001001000000111010001
    110010100100000011001100111010101100011011010110010000001100001011100100110111001
    010000001111001011011110111010100100000011011000110100101110011011101000110010101
    110100101101110011001110010000001110100011011110010000001101101011110010010000001
    111001001100101011100110110010101101110011101000110000101110100011010010110111101
    011111100100000010000010110110100100000011010010010000001110100011010000110000101
    010000001100010011011110111001001101001101110011001110011111110011100100000011000
    101011011100111010000100000011000100110000101101100011011000111001100100000011110
    111011101010010000001100011011000010110111000100000011100100110010101100001011001
    000011101000110100001101001011100110011111100100000011101110110100001111001001000
    100011010000110010100100000011001100111010101100011011010110010000001100001011100
    110011101000010000001111001011011110111010100100000011011000110100101110011011101
    101011011100110100101101110011001110010000001110100011011110010000001101101011110
    000011100000111001001100101011100110110010101101110011101000110000101110100011010
    111011011100011111100100000010000010110110100100000011010010010000001110100011010
    100011001000010111010000100000011000100110111101110010011010010110111001100111001
    110011001110101011000110110101101101001011011100110011100100000011000110111010101
    111010000100000011000100110000101101100011011000111001100100000011110010110111101
    010000001100011011000010110111000100000011100100110010101100001011001000010000001
    110100001101001011100110011111100100000011101110110100001111001001000000111010001
    110010100100000011001100111010101100011011010110010000001100001011100100110111001
    Partition you data
    Spread out you data
    Cache you data
    

    View Slide

28. 5
    E
    web scale
    we’ve got to be
    

    View Slide

29. web scale
    cloud scale
    what does that
    even mean?
    

    View Slide

30. twitter uses it
    but...
    it must be g d
    

    View Slide

31. choice of technology
    is not
    a function of fashion
    

    View Slide

32. it ok t be relational
    if it works for you & your data
    

    View Slide

33. 6
    E
    go daddy
    goes down
    

    View Slide

34. DNS is
    a single point
    of failure
    avoiding this is
    a tricky problem
    

    View Slide

35. have a simple back up plan
    Use more
    than one domain
    

    View Slide

36. 7
    E
    slow
    our app is
    

    View Slide

37. hogging resources
    100% CPU
    blocking
    locking
    opening cursors
    never closing
    interrupt
    obstructing
    stalling
    no connection pooling
    n+1
    leaking memory
    performance ++
    deadlocks
    

    View Slide

38. Just scale Up
    not optimised
    performance
    for node
    

    View Slide

39. Just scale out
    not optimised
    for cost
    

    View Slide

40. Bo leneck cost
    performance tuning is important
    

    View Slide

41. 8
    E
    chaos monkey
    

    View Slide

42. patche
    upgrade
    VM shuffling
    DNS changes
    shared hardware
    slo node
    Storage Transient failur
    queue Transient failures
    physical location
    time
    Storage Transient failures patches
    o upgrade
    queue Transient failures
    provisioning failure
    autoscaling
    capacity
    latency
    

    View Slide

43. patche
    upgrade
    VM shuffling
    DNS changes
    shared hardware
    Slo node
    Storage Transient failur
    queue Transient failures
    physical location
    time
    Storage Transient failures patches
    o upgrade
    queue Transient failures
    provisioning failure
    autoscaling
    capacity
    latency
    VOLATILE
    

    View Slide

44. a ept that
    It will fail
    and plan fo it
    

    View Slide

45. simulate chao
    get that monkey working for you
    

    View Slide

46. 9
    E
    you have lupus
    

    View Slide

47. it’s 2 am
    the phone rings
    your app is crashing
    servers are
    unresponsive
    

    View Slide

48. and it gets worse YOU CAN’T TELL
    WHAT IS GOING ON
    

    View Slide

49. centrally
    store logs
    

    View Slide

50. user activity
    correlate
    be able to
    

    View Slide

51. diagnostics
    n d t be a
    first class concern
    

    View Slide

52. 10
    E
    you are going to fail
    

    View Slide

53. accept it
    You’re not going to get it right
    not the first time
    and maybe not the time after that
    

    View Slide

54. change
    i coming
    

    View Slide

55. your designs responsibly
    to adapt
    be prepared
    

    View Slide

56. 10
    E
    phew
    thats it
    

    View Slide

57. for listening
    thanks
    @ammeep
    

    View Slide