This document discusses multi-cloud migration patterns and architectural strategies, highlighting the challenges and benefits of deploying applications across different cloud environments. It emphasizes the importance of planning and selecting appropriate migration patterns to maintain resilience, minimize code changes, and optimize resource usage. Various objectives and patterns for cloud migration are presented, along with a personal reflection on an Azure incident experienced by the author.

1. Cloud Migration Patterns: A Multi-Cloud
Architectural Perspective
Pooyan Jamshidi
Supervisor: Dr. Claus Pahl
IC4- Irish Centre for Cloud Computing and Commerce
School of Computing, Dublin City University
Pooyan.jamshidi@computing.dcu.ie

3. Source: cloudtweaks.com
P. Jamshidi, A. Ahmad, C. Pahl, "Cloud Migration Research: A
Systematic Review", IEEE Transactions on Cloud Computing,
2013, DOI:10.1109/TCC.2013.10.

5. Stairway to Heaven: An Architecture-Level Characterization of
Cloud Migration Strategies
Application Migration Patterns for the Service Oriented Cloud

6. N. Grozev and R. Buyya. Inter-cloud architectures and
application brokering: taxonomy and survey. Software:
Practice and Experience,

7. The Multi-Cloud Deployment is effective in dealing
with the following challenges:
• Users are not clustered near any single data center,
but form clusters around multiple data centers or are
widely distributed geographically
• Regulations limit options for storing data in specific
data centers
• Circumstances require that the public cloud be used
in concert with on-premises resources
• Application must be resilient to the loss of a single
data center

9. Azure incident that I personally experienced
…And it lasted for 6 odd hours!

10. …And it happened again in the
evening on the same day!

13. We do not intend to provide cloud application
architectural patterns. We rather abstract these
architectural layers into general concept of
components and connectors. The reason behind this
choice is that we have numerous cloud
architectures that becomes orthogonal to the
migration patterns and just add more ambiguity
into that.
Looking for cloud
application architecture?
Read these books

45. Objective MP1 MP2 MP3 MP4 MP5 MP6 MP7 MP8 MP9 MP10 MP11 MP12 MP13 MP14 MP15
Time to market -- -- -- -- --
New capabilities -- -- -- --
Reduce operational cost -- -- -- -- -- --
Leverage investments -- -- --
Free up on-premise resources
Scalability -- -- -- -- -- -- --
Operational efficiency -- -- -- -- -- -- -- --
Re-host
Cloudification
Relocation
Refactor
Rebinding
Replacement
Modernization

46. Objective MP1 MP2 MP3 MP4 MP5 MP6 MP7 MP8 MP9 MP10 MP11 MP12 MP13 MP14 MP15
Time to market -- -- -- -- --
New capabilities -- -- -- -- --
Reduce operational cost -- -- -- -- -- -- --
Leverage investments -- -- -- -- -- --
Free up on premise resources
Scalability -- -- -- -- -- -- --
Operational efficiency -- -- -- -- -- -- -- --

47. Objective MP1 MP2 MP3 MP4 MP5 MP6 MP7 MP8 MP9 MP10 MP11 MP12 MP13 MP14 MP15
Time to market -- -- -- -- --
New capabilities -- -- -- -- --
Reduce operational cost -- -- -- -- -- -- --
Leverage investments -- -- -- -- -- --
Free up on premise resources
Scalability -- -- -- -- -- -- --
Operational efficiency -- -- -- -- -- -- -- --

49. P. Jamshidi, A. Ahmad, C. Pahl, "Cloud Migration Research: A Systematic Review", IEEE Transactions on Cloud
Computing, 2013, DOI:10.1109/TCC.2013.10.
Migration patterns
Primarily facilitate
planning
Enable Auto-Deployment of
Multi-Tier Systems
in Hybrid Cloud Environment

52. Application Architecture (on-premise deployment)

53. Migration
Step
Requirement
Chosen
Patterns
1 Minimal code changes to application and familiarity with platform MP1
2 Granular control of resource usage and opportunity for auto-scaling MP6
3 Lower cost although some limitations on feature availability MP13
4 Replacing on-premise storage with cloud offerings MP12
5 Integration with cloud utility services MP14
6 Highly available service replacement MP11
7 Better user experience, improved efficiency, and load leveling MP3

54. Concern: Minimal code changes to application and
familiarity with platform
Selected pattern:
MP1

55. Concern: Granular control of resource usage and
opportunity for auto-scaling
Selected pattern:
MP6

56. Concern: Lower cost although some limitations on
feature availability
Selected pattern:
MP13

57. Concern: Replacing on-premise storage with cloud
offerings
Selected pattern:
MP12

58. Concern: Integration with cloud utility services Selected pattern:
MP14

59. Concern: Highly available service replacement Selected pattern:
MP11

60. Concern: Better user experience, improved
efficiency, and load leveling
Selected pattern:
MP3

62. Incremental migration
New functionalities by cloud offerings
Hybrid deployment
Multi-tenant architectures
Elastic software
SaaS

63. On-Premise Deployment
Multi-Cloud Deployment

64. Concern: Minimal code changes to application and
familiarity with platform
Selected pattern:
MP1

65. Concern: Granular control of resource usage and
opportunity for auto-scaling
Selected pattern:
MP6

66. Concern: Granular control of resource usage and
opportunity for auto-scaling
Selected pattern:
MP6

67. Concern: Enhancing availability without
architecture evolution and capital expenditure
Selected pattern:
MP4

68. Concern: solution is improved though best-in-
class cloud services
Selected pattern:
MP12

69. Concern: loose-coupled integration with cloud
utility services and re-hosted components
Selected pattern:
MP8

70. Concern: better user experience, improved
efficiency, and load leveling
Selected pattern:
MP3, MP6

71. Concern: business expansion and BPaaS
Selected pattern:
MP15

72. Incremental migration
Business expansion
Hybrid deployment
Multi-tenant business processes
Elastic business processes
BPaaS

74. The Message!

75. Pooyan.jamshidi@computing.dcu.ieI would be grateful to receive your thoughts on this: