Metron Athene Cloud Computing and Capacity Management

1. Metron AtheneCloud Computing and Capacity Management Leading the way in Capacity and Performance Management Charles Johnson – Principal Consultant www.metron-athene.com

2. Agenda What is Cloud Computing? What is Capacity Management? Cloud Capacity Management Challenges Application Transaction Management Reporting and Forecasting Summary www.metron-athene.com

3. What is Cloud Computing? To be considered a cloud . . . • On demand self service “A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service’s provider.” • Resource Pooling “The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand.” • Rapid Elasticity “Capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.”

4. What is Cloud Computing?

5. What is Cloud Computing? Cloud Ownership (or Deployment Models) • Public / External • Available to General Public • Owned by Organization selling Cloud Services • e.g. Amazon or Google • Private / Internal • Solely for organization • Onsite or hosted by 3rd Party • Community • shared by several organizations and supports a specific community that has shared concerns , i.e. Policy or Compliance • Managed either on or offsite by organizations or 3rd Party • Hybrid • composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load-balancing between clouds).

6. What is Cloud Computing? Service Models HaaS/IaaS – Hardware or Infrastructure as a Service PaaS – Platform as a Service SaaS – Software as a Service AasS – Application as a Service Maas – Monitoring as a Service Consumption-based / Subscription-based pricing

7. Popular Public Cloud Applications

8. What is Capacity Management? Capacity Management is Business Decision Support Reduces risk of demand exceeding available resources Protects ROI Every infrastructure change is a business decision Every infrastructure change has a cost impact Capacity Management equals Business Decisions www.metron-athene.com

9. What is Capacity Management? Capacity Management is • Strategic and Proactive • Assists in managing service quality • Assists in managing cost expenditures • Aligns business and IT • Match needs and cost during growth www.metron-athene.com

10. Cloud Capacity Management Challenges Capacity Management and the Cloud www.metron-athene.com

11. Forrester: Cloud needs Capacity Management Capacity management is the top operational concern--it's also a mystery as the underlying capacity of virtualized servers in public clouds is largely unknown. www.metron-athene.com

12. Cloud Capacity Management Challenges • Capacity on Demand - Too Little, Too Much or Just Right? • Automation and Control • Protecting your ‘loved ones’ • Scalability - In / Up or Out? • Old Habits – The Potential Risks

13. Cloud Capacity Management Challenges Capacity on Demand • Too Little (under capacity) • Application Sizing not performed • Controlled by limits • Lower usage costs • Service Levels affected – by how much? • High impact on business? • Loss of service • SLA breach penalties

14. Cloud Capacity Management Challenges Capacity on Demand • Too Much (over capacity) • Gartner – “most organizations overprovision their infrastructure by at least 100%” • Unlimited access to resources • Service Levels unaffected • Costs – higher resource usage, software licensing • Impacts on other services • Poor performance • Wasted resources (multi virtual CPU (vSMP) & single-threaded applications) • VM Sprawl • Increased pressure to manage virtual resources

15. Cloud Capacity Management Challenges Capacity on Demand • Just Right? • Application Sizing performed • Efficient use of IaaS • Acceptable Service Levels • continuous review and adjustment • Controlled by shares, limits and reservations • Continuous monitoring and tuning • Configuration adjustments (CPU, Memory) • VM consolidation • Power off unused ESX hosts • Right Balance • Find the equilibrium (service / cost)

16. Cloud Capacity Management Challenges Find the right balance

17. Cloud Capacity Management Challenges Automation and Control • Rapid Elasticity - Guest Migration and Portability, Templates, Golden Host • Resource Pools (limits, shares and reservations) • Load Balancing • DRS (Automatic, Partial or Manual) • Affinity • VM to VM or VM to Host • CPU • Fine grained tweaking for performance gains – Single Threaded Apps • Licensing

18. Cloud Capacity Management Challenges Protecting your ‘loved ones’ • Critical business applications • Service Levels must be met • Highest Priority (shares, unlimited), CPU affinity • Must guarantee resources (reservations) • High Availability Clusters • VMware - Fault Tolerance • Trade offs • “just in case” capacity management could impact on other services • Significant impact? Think about scaling out or up . . .

19. Cloud Capacity Management Challenges Scalability – Out For • Use existing hardware • Short lead time • No wait for new equipment • No need for hardware upgrade? • Costs are acceptable • SLA’s are met – but for how long? Against • Older and slower equipment • Not 64 bit, latest chipset technology • OK for the short term, but medium or long term? • Additional Costs • Licensing • Extended hardware support > 3 years • Power consumption and space

20. Cloud Capacity Challenges Scaling Up?

21. Cloud Capacity Management Challenges Scalability – Up? For • Upgrade to latest technology • faster CPUs / more memory • 64 bit compatible, hyperthreading enabled • Consolidation (more workload) • Greener • Efficiency Savings • All may outweigh scaling out to older systems Against • Longer lead time • Procurement of new equipment • Decommission old servers • Costs • Hardware Upgrade • Licensing

22. Cloud Capacity Management Challenges More things to measure – rapid response to new needs Correct sizing – minimize ‘cloudbursts’ Compare performance against SLAs Constant reassessment of provisioning requirements Evaluate provider against ‘good practice’ Private Public www.metron-athene.com

23. Cloud Capacity Management Challenges Scalability –In? • Spare capacity? • Sweat the assets by consolidating workloads • Use periods of inactivity • Migrate workloads between hosts • Turn off unused hosts • E.g. VMware DPM • Cost savings • Reduce Power • Reduce licenses

24. Cloud Capacity Management Challenges Old Habits – Potential Risks • Gartner – “Through 2015, more than 70% of private cloud implementations will fail to deliver operational, energy and environmental efficiencies.” • Infrastructure or application hugging • Silo mentality “what’s mine is mine” • Lack of resource sharing • Through lack of trust and confidence • “Just in case” capacity planning • Leads to over provisioning

25. Application Transaction Management Cloud Computing and Business Application Transactions Must understand the application transactions end-to-end Are service levels are being met?

26. Application Transaction Management Cloud Aware! www.metron-athene.com

27. Application Transaction Management Partner App Cloud Web Services Partner App App Servers Virtual Web Servers Firewall Wire Xfer <$100K Wire Xfer >$100K Msg. Broker Load Balancer Mainframe App Servers Proxy DB Server Virtual Web Servers Directory Server Network Firewall Problem – Ensuring Reliable IT Services Solution – Dynamic Transaction Path Detection Application Transaction Management

28. Application Transaction Management • Tracking transactions through all hops

29. Reporting and Forecasting • Implement Automated Reporting & Alerting • What do we need? • What recurrence? • Application and component • RAG Thresholds – SLA information (capacity, availability, response time) • Forecasting – Trending or Analytical Modeling or both? Make effective use of information . . . . . . .

30. Reporting and Forecasting • Which servers are hosting which applications? • Understanding the relationships • Understanding the volumes • Understanding the resource consumption • Accurately assigning resource usage

31. Summary Cloud Computing provides agility / elasticity to manage resources Capacity Management is critical to ensure Service Levels are met Capacity Management for Cloud environments is critical to ensure the efficient use of resources Application Transaction Management for Cloud applications is a must to become successful

32. Summary • Apply same Capacity Management principles • Gather the necessary information • Business – application sizing, usage periods, growth numbers • Service – what are the SLR’s, SLA’s, how do they impact on the business? • Component – gather and store the data • configuration, performance, application

33. Summary • Business runs on transactions, transactions run on IT • Transactions are the missing link between • Infrastructure • Applications • Business • Fully track transactions to optimize capacity management • Valuable business/IT data is created • Unprecedented visibility is gained • Enabling the use of this valuable data to drive better capacity management decisions and predictions