Navigating the Complexity of Distributed Microservices across AWS, Azure, and Google Cloud
- 1. Navigating the Complexity
of Distributed Microservices
across AWS, Azure, and
Google Cloud
Torsten Volk
Managing Research Director
Enterprise Management Associates (EMA)
Jasper Paul
Principal Product Manager
ManageEngine Site24x7
- 2. | @ema_research 2
Watch the On-Demand Webinar
• Navigating the Complexity of Distributed Microservices across AWS,
Azure, and Google Cloud On-Demand Webinar:
https://info.enterprisemanagement.com/navigating-the-complexity-
of-distributed-microservices-webinar-ws
• Check out upcoming webinars from EMA here:
https://www.enterprisemanagement.com/freeResearch
© 2023 Enterprise Management Associates,
Inc.
- 3. | @ema_research
Featured Speakers
With over 16 years of enterprise IT experience, Torsten
helps end users and vendors leverage the
opportunities presented by today's hybrid cloud and
software-defined infrastructure environments in
combination with advanced machine learning. He
specializes in topics that lead the way from hybrid
cloud and the SDDC toward a business-defined
concept of enterprise IT.
As a Principal Product Manager at ManageEngine
Site24x7, Jasper is passionate about all things AIOps,
observability, cloud infrastructure, and FinOps.
Jasper's tech expertise seamlessly merges with his
financial acumen to optimize IT operations. Be it IT or
life, Jasper finds joy in solving unique challenges with
a smile.
© 2023 Enterprise Management Associates, Inc. 3
Torsten Volk
Managing Research Director
Enterprise Management Associates
Jasper Paul
Principal Product Manager
ManageEngine Site24x7
- 5. Unfiltered
Complexity
Same data as title page:
2,932 Posts from
StackOverflow
25 out of 1040 problem
categories
Boxes sized by ‘number of
searches’
Result:
‘Infinite’ combinations
- 6. Modern App
Stacks Are
Complex
12 Layers (pink)
X
4-7 Components per
layer (blue)
X
Multiple choices per
component
= Complexity
And that’s just one
microservice
- 7. Distributed Apps: App Stack X 25
1. 25 codebases instead of one
2. 25 stacks instead of one
3. 25 release schedules instead of one
4. API calls instead of direct function call
- 10. Example: Networking Telemetry Data from One Microservice
Disjointed telemetry
data from the
networking layer
Data points are not
connected to
application code
or business
processes
And remember:
There are 11 more
layers
- 11. Data Driven Decision
Making Needs
Context
Critical Context Factors
- Who will be impacted?
- What is the extent of the
impact?
- What is the cost?
- How does all this affect key
business metrics?
- 12. Connecting the
Dots Between
Technology and
Business Is Key
The impact of each factor
has to be understood for an
observability platform to be
able to optimize
performance, resiliency,
cost, compliance, and
security.
- 15. 8 Ways of How AI Can Help Connecting the Dots
Between Infrastructure, Application, and Business
- 17. Observability + FinOps
Challenges
-Long time to parse Cloud bills
- Unable to identify cloud cost
leakages
- Difficult downloading Reports
- Multi-currency support at
Business Unit level
Actions
-BU visibility of cloud
-Auto remediate with IT
Automation with cloud monitoring
-Right size recommendations
-Integration with overall
ecosystem
- 18. Key Takeaways
Do not allow gaps in your telemetry data. Continuously scan for new
but unmonitored devices and services.
Always capture, analyze, and store data within its context.
Leverage AI for automatic root-cause analysis and optimal decision
making. Observability platforms need to deliver actionable insights
instead of disconnected alerts and dashboards.
Leverage FinOps to collaboratively optimize cost without sacrificing
resiliency, compliance, performance or security.