Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Adaptive Server Farms for the Data Center


Published on

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

Adaptive Server Farms for the Data Center

  1. 1. Adaptive Server Farms for the Data Center Contact: Ron Sheen Fujitsu Siemens Computers, Inc [email_address] Sever Blade Summit, Getting the most from your blade servers, March 22, 2005
  2. 2. The challenge – over provisioned, inflexible infrastructure <ul><li>Today‘s computing infrastructure – Static and unshared islands for each service </li></ul><ul><li>Inefficient </li></ul><ul><li>Costly </li></ul><ul><li>Over provisioned </li></ul><ul><li>Hard to manage </li></ul><ul><li>Inflexible </li></ul>Utilization of UNIX/Windows servers is low (<25% over 24 hours across all servers) Source: 2003 META Group – The Data Center of the Future
  3. 3. The challenge – Cost Reduction <ul><li>Labor costs - 75% of IT budgets - inefficiently applied across multiple, separate systems – mostly sustaining costs </li></ul><ul><ul><li>Excessive administration, upgrades, support, service, overhead </li></ul></ul><ul><li>Administration and routine IT tasks are mostly manual </li></ul><ul><ul><li>Limited labor resources/budget focused on manual, low value tasks </li></ul></ul><ul><li>IT over invests in resources for static, standalone systems </li></ul><ul><ul><li>Provisioning to meet peak demand </li></ul></ul><ul><ul><li>Provisioning for backup systems and applications </li></ul></ul><ul><ul><li>Resources duplicated; cannot be efficiently shared </li></ul></ul>
  4. 4. Addressing these challenges <ul><li>The tools exist to address these challenges by: </li></ul><ul><ul><li>Restructuring the data center to remove the barriers that isolate the various services into their various islands </li></ul></ul><ul><ul><li>Add tools to provision and monitor services and hardware </li></ul></ul><ul><ul><li>Add automation to automatically adapt to the changing needs of the services and users. </li></ul></ul>
  5. 5. Pooling and sharing of the overall resource <ul><li>Remove Boundaries </li></ul><ul><li>Consolidate Storage </li></ul><ul><li>Establish overall management </li></ul><ul><li>Assign Services </li></ul>Service A Service B Service C Service D Service E Service E Service B Service A Service C Service D
  6. 6. Automation for QoS, Scaling, HA, Configuration and Provisioning <ul><li>Assume a typical blade server farm - multiple blades/shared storage </li></ul><ul><li>Add server management, provisioning service, and control logic </li></ul><ul><li>Define policies and install monitoring and control agents </li></ul>Monitoring and control agents (A) provide information on application state and QoS Clients Shared Storage B1 B2 B3 Bn .… R/3 R/3 R/3 R/3 LAN2 (Storage) LAN1 (Public) Shared Storage Server Management And Provisioning Management Cluster Management Cluster A A A A
  7. 7. QoS Monitoring & Management High Water Mark Low Water Mark Target Metric Range Time QoS Metric Measured QoS metric exceeds the specified maximum acceptable value Allocate more satellite nodes and deploy needed application to meet QoS target Measured QoS metric is below the specified minimal acceptable value (too many resources) Perform orderly shutdown of some instances thus reducing cost and freeing the resources for other work.
  8. 8. Continuous Services Service B Adaptive infrastructure provides automated, continuous service and high availability without the costs of traditional infrastructure <ul><li>Application or server fails: </li></ul><ul><li>redirect user traffic …then </li></ul><ul><li>restart application or server </li></ul><ul><li>If no restart, but application service performance is satisfactory, there is no adjustment </li></ul><ul><li>If service performance is not satisfactory, then provision or reallocate another application/server </li></ul>Server Management And Provisioning Service E Service A Service C Service D Control Nodes Control Nodes
  9. 9. Summary of steps <ul><li>To implement adaptive server farms: </li></ul><ul><ul><li>Establish an infrastructure with sharable server platforms, storage and networks </li></ul></ul><ul><ul><li>Implement a deployment service </li></ul></ul><ul><ul><ul><li>This can be a traditional deployment service like e.g. Altiris or Remote Deploy </li></ul></ul></ul><ul><ul><ul><li>Or a virtual server deployment like VMWare </li></ul></ul></ul><ul><ul><li>Provide a highly-available management platform </li></ul></ul><ul><ul><ul><li>Monitor and collect information from all servers </li></ul></ul></ul><ul><ul><ul><li>Correlate data and execute reaction scripts </li></ul></ul></ul><ul><ul><li>Include agents on each server to provide data for the management platform </li></ul></ul><ul><ul><ul><li>Direct agents for monitoring application availability and performance </li></ul></ul></ul><ul><ul><ul><li>Interfaces to application suite tools like Oracle Grid Manager </li></ul></ul></ul><ul><li>The following slides illustrate two example deployments with SAP suite and Oracle. </li></ul>
  10. 10. Cost Reduction – Solution with server consolidation and automation Svc D <ul><li>Unshared </li></ul><ul><li>Utilization 10% </li></ul><ul><li>Peak 50% </li></ul><ul><li>Different peaks </li></ul><ul><li>Reduced investment </li></ul><ul><li>Higher utilization </li></ul><ul><li>Automation saves labor cost </li></ul>Spares Before After: Application Service Pools Spares with Automated Provisioning Service A Service C Service D Service B Service A Service B Svc C Control Systems
  11. 11. Server Consolidation example - SAP tech-1 Sample configuration for a SAP test environment Images SAP App SAP CI SAP Web WIN SQL Sample configuration for a SAP prod. environment SAP Web SAP Web SAP Appl SAP Central Spare WIN SQL SAP Web SAP Web SAP Appl SAP Central Spare WIN SQL SAP Web SAP Web SAP Appl SAP Central Spare WIN SQL SAP Web SAP Web SAP Appl SAP Appl SAP Appl SAP Appl
  12. 12. Server Consolidation example - Oracle Images Oracle App S Oracle RAC Web Cache & J2EE Oracle RDBMS SAP Web SAP Web SAP Appl Oracle RAC Spare Oracle RDBMS Web CachJ2EE SAP Web SAP Appl SAP Central Spare WIN SQL SAP Web SAP Web Oracle APP Svr SAP Central Spare WIN SQL SAP Web SAP Web SAP Appl SAP Appl SAP Appl SAP Appl Oracle front end environment Oracle decision support environment OLTP application environment Oracle RAC Oracle RAC Oracle RAC Oracle RAC Oracle RAC Oracle RAC Oracle RAC Oracle APP Svr Oracle APP Svr Oracle APP Svr Oracle APP Svr Oracle APP Svr Web CachJ2EE Web CachJ2EE Web CachJ2EE Web CachJ2EE Web CachJ2EE Spare Spare Spare Spare
  13. 13. Automated Server Farm Management <ul><ul><li>Dynamic service provisioning and workload management for blade server farms </li></ul></ul><ul><ul><ul><li>Automated, mass installation of bare-metal blade servers </li></ul></ul></ul><ul><ul><ul><li>Automated, mass software deployment and software updates </li></ul></ul></ul><ul><ul><ul><li>Priority and workload-based reusage of resources </li></ul></ul></ul><ul><ul><ul><li>Continuous services </li></ul></ul></ul><ul><ul><li>Benefits </li></ul></ul><ul><ul><ul><li>Reduced administration </li></ul></ul></ul><ul><ul><ul><li>Low-cost scalability </li></ul></ul></ul><ul><ul><ul><li>Simplified continuous availability </li></ul></ul></ul>Images Linux + sendmail W2K + IIS Linux + Apache W2K + Citrix + MS Office Linux+Apache Linux+Apache Linux+SendMail Linux+sendMail Linux+Apache Win+IIS Win+IIS Spare Win+Citrix Win+Citrix tech-1
  14. 14. Conclusion <ul><li>Many of the costly operations in the data center can be automated </li></ul><ul><li>These techniques will provide high available and high quality of service without the overhead and complexity of traditional clustering </li></ul><ul><li>The following demo will illustrate how this all works together. </li></ul>