5 lessons the “Cloud” can teach us about running more Efficient Environments

  • 1,499 views
Uploaded on

Dell global marketing director Andy Rhodes gave this presentation at Hostingcon. He talks about the 5 lessons he's learned from working with hyperscale customers about TCO and how these lessons can …

Dell global marketing director Andy Rhodes gave this presentation at Hostingcon. He talks about the 5 lessons he's learned from working with hyperscale customers about TCO and how these lessons can be applied to large scale hosters and IT environments.

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
No Downloads

Views

Total Views
1,499
On Slideshare
0
From Embeds
0
Number of Embeds
2

Actions

Shares
Downloads
0
Comments
0
Likes
2

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • For the operators of today’s clouds: search engines, social networking sites, online gaming platforms – the datacenter is the factory. The efficiency of it’s operation has direct impact on their financial results. These customers are highly aware of the cost drivers in their operations and Dell created the Data Center Solutions division to supply them with competitively advantaged, optimized solutions to lower their total cost of computing.Customers are asking for: Facility-level TCO. Velocity. Scaling. Globility. Requires a new approach that DCS was created to deliver.

Transcript

  • 1. 5 lessons the “Cloud” can teach us about running more Efficient Environments
    Andy Rhodes
    Director: Datacenter Solutions
  • 2. Facts on Dell in the Cloud…THIS IS NOT A SALES PITCH!
    2
  • 3. The Datacenter is the Factory
    3
  • 4. 4
    The Largest Clouds in the World want Custom Infrastructure
  • 5. DCS case studies
    5
  • 6. Lesson Number 1: Only Your TCO (Total Cost of Ownership) model is right… but everyone can contribute to it
    6
  • 7. Lies, damn, lies and TCO models Generic TCO models often created by vendors FOR vendors. Enterprise Apps have different models than Cloud/Large scaled out solutions.Don’t fool yourselves…. Get EVERY cost in there(Remember this business is now about cents versus dollars!)
    7
  • 8. Sample of a real life TCO model
    8
  • 9. 9
    Our Advice: Build a Robust TCO model
    geared to your business model but
    flexible enough to accommodate new information and market changes
  • 10. Lesson Number 2: Don’t let the Status Quo hold you back
    10
  • 11. Industries often take time to evolve, look at cutting edge vendors with radical designs (and derivatives thereof…)
    11
  • 12. 12
  • 13. 13
    Real Savings
    • Customer Charged $4/Watt/Year to Operate Server in Data Center
    • 14. Charge of $40 per U space saved
    • 15. 3 year TCO cycle
    • 16. Deploying 1000 servers
    • 17. PUE of 1.4
    • 18. Double Density
    • 19. 65 Watts less power
    1U Monolithic
    Unoptimized
    Optimized 4 in 2U
    $780,000 Operating Costs saved
    Mega-Watts Reduction
  • 20. Apply that thinking to services as well..
    14
  • 21. Don’t always drink the Kool Aid. Work those designs into your TCO model
    15
  • 22. Lesson Number 3: The most expensive server/storage node is the one that isn’t used (sits idle for 6-12 weeks)or it’s the one you don’t have when you need it most!
    16
  • 23. 17
    Mega Scale Cloud Suppliers run Just in Time (J.I.T) IT shops. Working capital is not left idle on the factory floor
    We’re an on-demand & competitive society. If you’re not ready don’t expect your customers to wait around for you. (ours don’t….)
    Our Advice: Capacity Planning is not just your problem, hold your partners accountable as much as your internal business units. (BTW account for this in your TCO model)
  • 24. Lesson Number 4: Don’t allow Bad Code to dictate the hardware architecture. The ones who do will be at a commercial disadvantage
    18
  • 25. Bad Code = belts and braces (suspenders)
    19
  • 26. Bad Code = more servers and storage nodes than necessary
    20
  • 27. Scaled out apps seriously reduce the number of system admin’s per infrastructure node. Strive for that model (even if its over time).
    21
    Revolutionary
    • “Leapfrog” Efficiency Gains
    • 28. Platform-based
    • 29. Application-oriented
    • 30. New benchmark of CIO’s
    • 31. Dell is powering 20 of 25 largest
    • 32. Incremental Efficiency Gains
    • 33. Virtualization-based
    • 34. Infrastructure-oriented
    Evolutionary
    Efficiency
    Open, Capable, Affordable
    • Inflexible Form-Factors
    • 35. Complex Layered Networking
    • 36. Tool/Process Lock-in
    • 37. Expensive, Proprietary HW
    Time
    Today
  • 38. Advice: buy base elements and innovate around your strengths. There are enough ‘cloud” stacks at the IaaS and PaaS level now that are robust and ready to deploy.
    22
  • 39. 23
    Lesson Number 5: Don’t waste time on “Cloud Washing”. Talk to your customers about real pain points and how to solve them
  • 40. Even if Cloud is more popular than Britney Spears…. Customers are past the basic “what is it” and want real solutions
    24
  • 41. Only Your TCO model is right… but everyone can contribute to it
    Don’t let the Status Quo hold you back
    The most expensive server/storage node is the one that isn’t used (sits idle for 6-12 weeks)
    Don’t allow Bad Code to dictate the hardware architecture. The ones who do will be at a commercial disadvantage
    Don’t waste time on “Cloud Washing”. Talk to your customers about real pain points and how to solve them
    25
  • 42. Thank youwww.dell.com/cloud