Founder's Perspective

571 views

Published on

Analyst Day 2011 presentations by Dave Hitz
Founder & EVP

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
571
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
7
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Founder's Perspective

  1. 1. Financial Analyst Day – June 30, 2011Founder’sPerspectivesDave HitzEVP and Founder
  2. 2. A Combo of Historic Proportion Value NetApp Delivered SATA Cost RAID-DP Reliability Flash Speed
  3. 3. A Combo of Historic Proportion Value EMC NetApp Delivered 20 Years Ago SATA Cost SCSI RAID-DP Reliability Mirroring Flash Speed RAM
  4. 4. A Combo of Historic Proportion Value EMC NetApp Delivered 20 Years Ago SATA Cost SCSI RAID-DP Reliability Mirroring Flash Speed RAM Single platform increases market leverage Key benefits:  Same performance, half the price  Configurable performance
  5. 5. A Scale Evolution Pattern Fast deployment  Spec’d by development, not IT  Do what’s easiest to get going Grow fast  Do more of what you did before  Things break: cost, performance, and manageability Enterprise-ready deployment  Cost & operational efficiency  Performance & reliability  Manageable, repeatable
  6. 6. A Scale Evolution Pattern Innovation Fast deployment  Spec’d by development, not IT  Do what’s easiest to get going Grow fast  Do more of what you did before  Things break: cost, performance, and manageability Innovation Enterprise-ready deployment  Cost & operational efficiency  Performance & reliability  Manageable, repeatable
  7. 7. Example: Networked Storage Direct Attached Storage (DAS)  Expensive  Inefficient  Complex Networked Storage
  8. 8. Example: Networked Storage Direct Attached Storage (DAS)  Expensive  Inefficient  Complex Networked Storage
  9. 9. Example: New Analytics orOne big, expensive Many small, cheap computer computers
  10. 10. Example: New Analytics  Database with Hadoop Map/Reduce  Lots of CPUs  Existing network & DAS  Performance scaling just works  Overwhelms network  Overwhelms storage Works when… Doesn’t work when… Uniform problem  Multiple problems; non-uniform data on uniform data  Workloads without balance in capacity/compute  Have to buy servers to add capacity  Have to buy storage to add compute
  11. 11. Example: New Analytics  Database with Hadoop Map/Reduce  Lots of CPUs  Existing network & DAS  Performance scaling just works  Overwhelms network  Overwhelms storage  Enterprise-ready software solutions  CPU farms  Faster, bigger networks  Hadoop networked storage solutions
  12. 12. Example: Flash/Storage on the Host CPU  Add storage to host  Unit of scale = Flash CPU + Storage
  13. 13. Scale drives IT markets toward: Efficiency Manageability Shared infrastructure

×