This session will explore what happens after server consolidation: Workload profiles change, usage patterns change, and resource utilization goes up and down. During this session, we will discuss how the storage footprint changes, and what used to be ideally optimized isn't anymore. You will learn how to quantify the gap between resource allocation and actual usage, how to recover and redeploy excess capacity, and how to find common virtual machine configuration mistakes that can degrade overall performance.

1. Finding Virtual Coins
in the Couch
Mike Mckay
mike.mckay@novell.com

2. Session Overview
Today's Data Center
Virtualization adoption rates
The over-provisioned and under-utilized virtual farm
PlateSpin Recon and the 4 Steps to Success
®
Awareness, analysis and planning
Virtual capacity management
Identification of future bottlenecks, opportunities for reclamation,
and configuration issues
PlateSpin Recon Demo: Finding Virtual Coins
2 © Novell, Inc. All rights reserved.

3. Today's Data Center

4. Today's Data Center
Most of the large and mid-tier enterprises are well
along the path to virtualization
Gartner Group stated in the December 2009 Data
Center conference “The Global 500 has
approximately 25% of x86 workloads virtualized”
2 common methods of implementing virtualization
Over provisioning of the virtual infrastructure
(large initial investment)
Pay as you go virtual infrastructure
(purchase more capacity (nodes) as needed)
4 © Novell, Inc. All rights reserved.

5. The Truth and Reality
(of the Virtualization team)
The Truth...
A virtualization teams time is spent performing
three (3) main tasks, day to day.
Creating new virtual machines for the business users
Day to day administration of the environment
Performing and continuing to migrate the remaining
physical workloads.
5 © Novell, Inc. All rights reserved.

6. The Truth and Reality
(of the Virtualization team)
The Reality...
Optimization of the new environment was NOT
their responsibility!
Today, as Gartner Group stated in the December
2009 Data Center conference “The Global 500 has
approx. 25% of x86 workloads virtualized, although
the fastest growth area is SMB.” “By 2012 their will
be approx. 58 Million virtual machines deployed”
(with lot's more to add).
6 © Novell, Inc. All rights reserved.

7. Challenges with Virtual Infrastructures
Am I getting the most ROI from my purchases?
Can I achieve higher consolidation ratios and
VM density?
Can I rearrange VMs to free up virtual resources?
Do I have more server consolidation opportunities?
Which VMs are facing resource constraints?
When will I need to purchase more virtual hosts,
storage, etc.?
Are my VMs using their allocated capacity?
Can I right-size VMs to free up capacity?
7 © Novell, Inc. All rights reserved.

8. Goals for Data Center Optimization
Some examples:
Maximize data center efficiency
Increase VM performance
Find, reclaim and redeploy unused virtual capacity
Make more efficient use of existing resources
Postpone purchases of new resources
Predict future resource needs
8 © Novell, Inc. All rights reserved.

9. PlateSpin Recon Enterprise Overview
®

10. Awareness, Analysis and Planning
Data collection, analysis and reporting for the
What resources are
in the data center Data Center Manager and IT Architect
What workloads
are running on
those resources
How effectively are
workloads assigned
to resources
Planning and
scenario modeling
Workloads Resources
10 © Novell, Inc. All rights reserved.

11. Virtual Capacity Management
Maximize virtualization investments by understating
how your virtual resources are being utilized
Correct configuration issues and right-size server
resource allocation to better match the workload
Create more space in your virtual environment and
improve resource capacity
Reclaim unused virtual resources and
defer the cost of new server purchases
11 © Novell, Inc. All rights reserved.

12. Virtual Capacity Management
Awareness of
Available Resources
Identification Virtualization Opportunities
of Bottlenecks Capacity for Reclamation
Management
Configuration Issues
Cost Saving/Avoidance
12 © Novell, Inc. All rights reserved.

13. The Ideal Virtual Environment
Optimized Under/Over Utilized
13 © Novell, Inc. All rights reserved.

14. PlateSpin Recon: ®
Understand Resource Utilization
1
Collect VM utilization
2
Identify bottlenecks
3
Right size and
data over a standard and areas of adjust VM resources
business cycle reclamation allocation to improve
(30 days) the performance
of available services.
Supply
Unused Server Resources
Demand
Virtual Machine
14 © Novell, Inc. All rights reserved.

15. Opportunities for Reclamation
Supply
Waste
Demand
Do I have resources assigned that aren’t being used?
Do I have VMs that aren’t being used?
Reduce waste, assign unused resources to new projects =
Defer the purchase of new hardware
15 © Novell, Inc. All rights reserved.

16. Opportunities for Reclamation
16 © Novell, Inc. All rights reserved.

17. Identification of Bottlenecks
Supply Supply Supply
Demand
+ Demand
=
Demand
When will aggregate supply exceed aggregate demand?
What resource (CPU, memory, network, disk) have I run out of/
will I run out of next?
Improve performance, satisfy service levels with existing hardware =
Defer the purchase of new hardware
17 © Novell, Inc. All rights reserved.

18. Identification of Bottlenecks
18 © Novell, Inc. All rights reserved.

19. Configuration Issues
10 Ways to Increase VMware ESX Server guest
Performance on a OS performance tips
VMware System Tip 2: Improve your host's
10. Use VMware Tools processor and memory
Tip 5: Be aware of
6. Disable the CDROM VM-to-host placement
in VMware
Tip 7: Virtual machine
2. Upgrade your CPU processors and memory
1. Upgrade your RAM Tip 8: Remove unneeded
(more RAM) virtual hardware
Source: www.petri.com Tip 9: Update VMware Tools
Source: searchvmware.techtarget.com
Improve performance, satisfy service levels with existing hardware =
Defer the purchase of new hardware
19 © Novell, Inc. All rights reserved.

20. Configuration Issues
20 © Novell, Inc. All rights reserved.

21. The 4 Steps to Success

22. The 4 Steps to Success
1
Install: PlateSpin Recon Enterprise
®
in the data center
Installed on dedicated collector servers
No agents
Data remains on site
22 © Novell, Inc. All rights reserved.

23. The 4 Steps to Success
2
Inventory: Discover and inventory
every physical server,
virtual host and virtual
machine in the data center
Server type, CPU type, number of cores
Name, IP address, domain
Operating system, patches,
hotfixes, applications
Services
Storage (total and available)
23 © Novell, Inc. All rights reserved.

24. The 4 Steps to Success
3
Monitor: The resource utilization of
each workload, physical server
and virtual host for 30 days
CPU utilization
Storage utilization
Memory utilization
Network throughput
Disk throughput
Complete 24-hour profile,
not just peak usage
24 © Novell, Inc. All rights reserved.

25. The 4 Steps to Success
4
Analyze and Report: Detailed reports
and analytics on actual and
projected system performance
Recognize underused virtual hosts
Pinpoint current capacity bottlenecks in virtual
hosts and predict future ones
Discover allocated but unused virtual resources
Identify mis configured virtual hosts and VMs
25 © Novell, Inc. All rights reserved.

26. The Next Steps:
Tracking and Billing the Virtual Environment

27. How to Successfully Manage Capacity
What administrators need to know:
Who is using your VMs?
What are they using your VMs for?
When will they be done with your VMs?
Where are your VMs assigned to run?
Develop standard VMs configurations, and identify
VMs that don’t conform
Charge users for their usage of the virtualization
environment, at a minimum track who is using how
much of your environment
27 © Novell, Inc. All rights reserved.

28. How to Successfully Manage Capacity
IT Chargeback is a way that IT organizations allocate their costs
to different business units based on usage of IT resources
PlateSpin Recon measures resource usage for all virtual machines
®
The Chargeback framework converts resource usage figures
into billing vales over set time periods
Network
Disk Data CPU Data
Data
Each machine can have multiple raters.
Raters can bu used for multiple machines.
Rater
28 © Novell, Inc. All rights reserved.

29. How to Successfully Manage Capacity
Raters associate resources (inventory data and/or monitored
metrics) with a fee that may or may not be time
dependent, examples:
Each processor used by the workload will cost $100
Each MB of memory used, based on average daily use, will cost $0.01
Workloads are then attached to raters as applicable, examples:
This workload will be attached to the “Processor Assigned” rater
These Workloads will be attached to both the “Processor Assigned” and
“Memory Used” raters
Reports are then run to calculate the cost for each workload
over a selected time period, examples:
This workload costs $200 for the last month
These workloads costs $300 for the last week
29 © Novell, Inc. All rights reserved.

30. PlateSpin Recon Demo:
®
Finding Virtual Coins

31. Demo Overview
Inventory and Monitoring
Virtualization adoption rates
The over-provisioned and under-utilized virtual farm
Finding Virtual Coins in the Couch
Inventory reports
Reclamation opportunities report
Bottleneck identification report
Configuration optimization report
The Next Steps: Tracking and ChargeBack
31 © Novell, Inc. All rights reserved.

33. Unpublished Work of Novell, Inc. All Rights Reserved.
This work is an unpublished work and contains confidential, proprietary, and trade secret information of Novell, Inc.
Access to this work is restricted to Novell employees who have a need to know to perform tasks within the scope
of their assignments. No part of this work may be practiced, performed, copied, distributed, revised, modified,
translated, abridged, condensed, expanded, collected, or adapted without the prior written consent of Novell, Inc.
Any use or exploitation of this work without authorization could subject the perpetrator to criminal and civil liability.
General Disclaimer
This document is not to be construed as a promise by any participating company to develop, deliver, or market a
product. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in
making purchasing decisions. Novell, Inc. makes no representations or warranties with respect to the contents
of this document, and specifically disclaims any express or implied warranties of merchantability or fitness for any
particular purpose. The development, release, and timing of features or functionality described for Novell products
remains at the sole discretion of Novell. Further, Novell, Inc. reserves the right to revise this document and to
make changes to its content, at any time, without obligation to notify any person or entity of such revisions or
changes. All Novell marks referenced in this presentation are trademarks or registered trademarks of Novell, Inc.
in the United States and other countries. All third-party trademarks are the property of their respective owners.