SUSE Conversations


Server Consolidation and Data Center Consolidation



By: battala

March 17, 2010 11:28 am

Reads:406

Comments:0

Rating:0

By Rajesh Battala

Server consolidation is a key element of cutting IT costs. Novell products make it easy to assess your data center for consolidation and optimization. This not only helps your IT organization to reduce costs but also improves the efficiency of the usage of the resources, improves service levels and helps in better information organization.

The purpose of server consolidation is to use your computer servers more efficiently so you can reduce the total number of physical servers or server locations. Server sprawl is one of the main reasons you need to worry about server consolidation.

Server consolidation benefits include:

  • Reduce Hardware Costs
  • Reduce Maintenance Costs
  • Reduce Support Costs
  • Reduce Floor Space & Facility Costs.

Several factors typically drive an organization to look at server consolidation:

  • As business becomes more competitive and recession affects the market, enterprises need to reduce their costs by using existing resources more effectively.
  • Advancements in hardware from Intel and AMD have increased their capabilities with inexpensive multiprocessor and 64-bit machines.
  • Virtualization support from some hardware vendors — Intel’s-VT capability and AMD-V Virtualization — makes virtualization more feasible.

There are lots of solutions available in the market for data center management. Novell’s Virtualization and Workload Management solutions are particularly well suited to help your organization with server and data center consolidation.

Platespin Recon and Platespin Migrate are two products from Novell which can help you implement server consolidation and data center management with ease.

To Implement Server Consolidation:

  1. Inventory resources in the data center
  2. Monitor the resources’ usage and capabilities
  3. Analyze results
  4. Determine a server consolidation plan
  5. Execute the consolidation
  6. Manage the consolidated servers

Here is how administrators can execute the above steps using Novell products.

How PlateSpin Recon helps Server Consolidation:

PlateSpin Recon is not only a good monitoring product but also a great server consolidation reporter. It collects workload inventory and resource usage statistics.

PlateSpin Recon supports a high number of platforms.

PlateSpin Recon Platform Support

Windows Linux Novell UNIX Hypervisors
  • Windows 2008 Server (32 and 64-bit)
  • Windows Vista (32 and 64-bit)
  • Windows 2003 Server (32 and 64-bit)
  • Windows XP pro
  • Windows 2000
  • Windows NT 4.0
  • SUSE Linux Enterprise (32 and 64-bit)
  • Open SUSE (32 and 64-bit)
  • Red Hat Linux (32 and 64-bit)
  • Fedora (32 and 64-bit)
  • CentOS (32 and 64-bit)
  • Ubuntu (32 and 64-bit)
  • NetWare
  • Solaris (32 and 64-bit) SPARC
  • AIX¹
  • VMware Infrastructure
  • VMware ESX
  • VMware ESXi
  • VMware Server
  • Microsoft Hyper-V
  • Microsoft Virtual Server
  • SUSE Linux Enterprise Server with Xen
  • Citrix XenServer
  • Virtual Iron
  • Solaris Containers
NOTE: for more information on platform support see: http://www.novell.com/products/recon/techspecs.html

The following steps to implementing server consolidation are made a lot easier with PlateSpin Recon:

  1. Inventory the resources/workloads in the Data Center.
  2. Monitor the workloads for a period of time.
  3. Analyze the workloads’ usage from their monitoring data.
  4. Identify the candidates for consolidation.
  5. Generate a server consolidation plan.

1. Inventory the resources\workloads in the Data Center.

To inventory the resources, Recon supports

  1. Individual machines
  2. Virtual Center
  3. IP Range
  4. Windows Domain
  5. CVS File
  6. SNMP
  7. subnet

To inventory the machines we need to provide the credentials of the machines. PlateSpin Recon provides a good and easy way to manage the credentials of machines and maintain them securely.
Consider a case where you need to inventory the machines using IP Range Mechanism.

You just create some credential sets and assign them to particular platforms. Recon uses these specified credential sets to authenticate all machines in that IP range and inventory them. This will avoid a lot of the headaches of providing credentials to each machine.

If there is any failure in inventory just refer to the job logs which will help you understand why it is failing. Once the inventory of all the machines is successfully done, the next step is to Monitor the Machines.

2. Monitor the workloads for a period of time

To monitor the workloads of a machine, right-click and choose Start Monitoring. It gets the performance data of the server. To monitor a group of machines, select all the machines, right-click and choose Start Monitoring. At any time you can stop monitoring by right-clicking and choosing Stop Monitoring. Any time Monitoring fails, the associated log will be shown in the client which will provide necessary information for the problems.

Recon collects several detailed metrics of a workload depending on the inventory type. For example, tracking detailed metrics on storage usage over time, versus a simple point-in-time snapshot, PlateSpin Recon shows how and when physical servers consume local and remote storage resources iSCSI and FC SAN (at both the physical disk and logical volume level).

3. Analyze the workloads usage from their monitoring data

Data must be collected at least one hour before running Analysis Reports. To view the monitoring data or usage of the resources, Recon provides a chart where the monitoring data of the machines will be shown graphically.

Reports:

PlateSpin Recon reporting will show the inventory and usage information in a detailed way.

Go to Report Explorer

Reports are organized according to category.

If you want to know all the inventory information, go to Inventory Node, select a category and right-click and generate a report. A report will be created and saved in the server.

4. Identify the candidates for consolidation

Generate the reports after monitoring the workloads. The reports will give a good summary of how the resources are getting used in the Data Center.

There are three Consolidation Candidate reports generated that contain a listing of good consolidation candidates, poor consolidation candidates, and candidates requiring further investigation.

After monitoring the workloads for a certain time, Recon reports which tell which are good and bad candidates for server consolidation.

Go to Report Explorer under Consolidation and Consolidation Project Samples, select the appropriate template and generate the report.

Recon can deliver the reports via email, save as Network File, FTP, and will show in report explorer.

It also supports a wide range of output formats (pdf, png, jpg, excel, csv).

The report can be scheduled or it can be recurring to keep the administrator fully aware of data center status.

5. A server consolidation plan

You should monitor the machines for at least 30 days to generate a good server consolidation plan. Consolidation scenarios use the performance data captured during monitoring to generate server consolidation scenarios.

The hardware, virtualization, and power and cooling information, as well as total cost of ownership and utilization thresholds, are values associated with PlateSpin Recon Server Templates.

Before starting on generating server consolidation reports, you should first determine the target hardware servers and virtualization technology.

Server Templates play a major role in server consolidation. By default there will be some commonly used server templates in Recon. Provisions are there to create or edit the server templates.

How to determine the consolidation and protection of workloads

  1. After monitoring the workloads for a period of time, go to Project Explorer and create a new project.
  2. Select the Project Type ( server consolidation or workload protection or both).

  3. Click Next and select all the machines for server consolidation.

  4. Click Next to specify the Data Range and the Business Days.

  5. Click Next to enter the Forecast Options.

  6. Click Next to specify the whether to Enable the Power and Cooling factors and their values.

  7. Now we are ready to create the project. Click Create to create the project.

Once the project is created it will be shown in the Project Explorer.

Under the project

Scenarios: user created scenarios will be shown
Existing Target Server: Contains the list of server class machines added to the project
Consolidate: workloads to be consolidate will be shown
Protect: workloads that can be protected will be shown

Some scenarios are:

  1. Determine the number of servers (selected from the server templates) required for server consolidation.
  2. User can specify the servers (from server templates or existing servers) and Recon will create a consolidate report accordingly.

In the first case, Recon will generate a report of how many specified servers are required to consolidate the workloads and how to group them also.

In the second case it will fit workloads well into the servers provided by the user.

The output of server consolidation reports are very optimal.

This is how PlateSpin Recon helps to generate the server consolidation reports to the administrator.

From this point PlateSpin Migrate takes control of implementing the scenarios generated using Recon.

PlateSpin Migrate

As businesses grow rapidly there is a rapid increase in the computer systems in terms of size and management. These computer systems will have different configurations such as operating system, applications running on them, and hardware capabilities.

PlateSpin Recon helps by providing reports about how these computer machines are being used in the data center. Implementing the server consolidation plans generated by Recon, depending on the users’ needs, helps you manage and protect these workloads.

Protecting or moving the workloads is not a simple task, as it involves lots of tasks to be done very carefully and sometimes without disruption of the workloads. PlateSpin Migrate safely and easily moves or protects the workloads in the data center.

Migration of the workloads is either moving the workload on physical machine to a virtual machine or from a virtual machine to physical machines.

Protection of the workloads is to take the backups of the workloads into different images, so when a particular workload gets corrupted or damaged, the protected workloads will start running to provide the services.

Inventory Supported by PlateSpin Migrate:

Server Operating Systems Desktop Operating Systems Hypervisors Hardware Vendors
(Standalone and Blade Servers)²
Image Formats
  • Windows 2008 (32 and 64-bit)
  • Windows 2003 (32 and 64-bit)
  • Windows 2000
  • Windows NT 4
  • SUSE Linux Enterprise Server (32 and 64-bit)¹
  • Red Hat Linux (32 and 64-bit)¹
  • Sun Solaris (Sparc)¹
  • Windows Vista (32 and 64-bit)
  • Windows XP pro
  • Windows 2000
  • VMware ESX
  • VMware ESXi
  • Microsoft Hyper-V
  • Microsoft Virtual Server
  • Virtual Iron
  • SUSE Linux Enterprise with Xen
  • Citrix XenServer
  • Solaris Containers
  • Dell
  • HP
  • IBM
  • Unisys
  • Sun
  • Others
  • Acronis
  • TrueImageTM
  • PlateSpin Flexible
  • Image Packages
  • Symantec GhostTM
  • Symantec LiveStateTM
  • CommVault
  • Tiyoli Storage Manager
  • Symantec Backup ExecTM

¹ Offline migration only
² Contact PlateSpin for current hardware compatibility
For more information on the support matrix please see: http://www.novell.com/products/migrate/techspecs.html

Conversions supported by PlateSpin Migrate

PlateSpin Migrate supports conversions across multiple physical to virtual infrastructures (and vice versa) and heterogeneous virtual infrastructures.

PlateSpin Migrate can do Machine to Physical (P2V and V2V), Physical to Machine (V2P and P2P), Capturing the Image of a workload (X2I) and Image Deployment.

PlateSpin Migrate will allow the user to customize the configuration of the target machine before it gets migrated. When the migration is over then the workload will be ready with the customized configuration and will be ready to provide the services.

PlateSpin Migrate can do migration of Windows/ Linux workloads across Physical to Virtual and Virtual to Physical, and across different virtual infrastructures like VMware, Citrix xen, Xen etc.

PlateSpin Migrate helps disaster recovery by the methods of copying/moving the workload to different image formats. The supported image formats are specified in the above table.

PlateSpin Migrate lets the administrators schedule the conversions after doing the configuration and customization.

Other Tools

PlateSpin Orchestrate

PlateSpin Orchestrate will manage the heterogeneous hypervisors and workloads well from a single point.

SUSE Studio

SUSE Studio is a Linux software appliance creation tool where the users can create appliances of different formats that are ready to deploy and use in the enterprise.

Conclusion

PlateSpin Recon will generate a optimal server consolidation and protection plans and PlateSpin Migrate implements the plans. This how Novell PlateSpin Recon and PlateSpin Migrate provides best solutions for server consolidation and data center consolidation.

More Articles about Server Consolidation – Data Center Consolidation

VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)

Tags: , ,
Categories: Enterprise Linux, SUSE Linux Enterprise Server, Technical Solutions

Disclaimer: As with everything else at SUSE Conversations, this content is definitely not supported by SUSE (so don't even think of calling Support if you try something and it blows up).  It was contributed by a community member and is published "as is." It seems to have worked for at least one person, and might work for you. But please be sure to test, test, test before you do anything drastic with it.

Comment

RSS