Ensuring Technical Readiness For Copilot in Microsoft 365
Cloud computing
1. Cloud Computing
Richard Astbury
Senior Consultant
two10 degrees
2. What is cloud computing?
• Compute and storage as a utility
Pay as you go, starting at zero
• Scalable
Infrastructure is elastic
3. Load Forecast
Allocated
Not Enough IT capacity
Power
IT Capacity
Too Much
Power
Actual Load
Time
4. Capacity on
No screaming customers
Demand
IT Capacity
No capital
laying idle
Goes up
and
down
Actual Load
Lower Capex Time
5. On and Off
Compute
Inactivity
Period
Average Usage
Growing Fast
Compute
Average Usage
Unpredictable Bursting
Compute
Average Usage
Compute
Predictable Bursting
6. What is cloud computing?
• Compute and storage as a utility
Pay as you go, starting at zero
• Scalable
Infrastructure is elastic
• Hosted in data centres around the globe
Microsoft / Amazon / Google
9. Why?
• Lower costs
Less responsibility
Offer customers new features
Reduce piracy
• If they don’t, someone else will!
10. A Golden Age
• Tooling is now free
Visual Studio / Eclipse etc…
• Explosion of 3rd party libraries
GitHub.com
• Help is readily available
StackOverflow.com
• Low barrier to entry
node.js / Ruby on Rails / Django
• Markets are now global
Viral marketing
11. Summary
• Building applications has never been so
easy
• The internet gives you a global market
• Cloud Computing is the platform for you to
deliver
13. Questions
Twitter @richorama
Web two10degrees.com
Email richard.astbury@two10degrees.com
Editor's Notes
Slide Objectives:Understand the challenges of irregular load in applicationsUnderstand the challenges of both too much capacity and not enough capacity and the sort of groups they impact within a customerUnderstand that traditionally IT capacity is purchased in a stepwise fashion involving capital expenditure at each pointSpeaking Points:To understand the value and opportunity for cloud computing, I believe it’s important to think about how IT capacity is typically used by applications today in most environments.Today most organizations significantly over estimate or underestimate the amount of resources they need to run their applications.This leads to a higher cost for the infrastructure and the delivery of the overall applications. Build Steps:Forecast load is to grow steadilyPlanned capacity grows in a stepwise fashion. Need to plan in advance due to hardware lead timesActual load is highly variable over timePeriods where we have excess capacity. Capital laying idle, opex wasted powering and cooling serversPeriods where we have insufficient capacity and our customers get a bad experience