User:Rikoherwanto/sandbox

Cloud computing offers exceptional flexibility and fantastic choices compared to traditional computing or storage methods. Many cloud services operate on a pay-as-you-go model that has no capacity or usage restrictions. The question becomes; is capacity planning needed in the cloud? Although at first glance planning may seem irrelevant, despite the "infinite" appeal, it is very important to keep using comprehensive planning when determining the capacity of cloud resources. Proper cloud planning can save budget, streamline IT operations and maximize efficiency. One of the biggest advantages offered by Cloud Computing is the use of resources in an elastic manner. Cloud requires less hardware than traditional computing structures which can mean greater flexibility and lower initial costs. Even better, it can be easy and fast to buy additional resources. However, elasticity is only good if the implementation strategy is good Conventional Capacity Planning Includes:
 * Facility considerations (Space, Power, Network, Cooling)
 * Supply Chain Management Constructions and Relationships
 * Hardware lifetime contour & failure rates (MTBF)
 * Systems management staff
 * Seasonal and unexpected burst considerations
 * Workload colocation and performance demands
 * Over-provisioning for reliability and rate of innovation
 * Effective tooling
 * Business continuity planning

(Cloud) Capacity Planning Covers
 * Seasonal and unexpected burst considerations
 * Workload colocation and performance demands
 * Over-provisioning for reliability and rate of innovation
 * Effective tooling
 * Business continuity planning

Here are some things that might be included as a checklist in planning cloud computing capacity.

ROI from transformation
Today, various organizations are constantly challenged to reduce their overall technology expenditure. Cloud computing is one of the key methods for transformation that organizations use to achieve this goal. Organizations that are assisted by cloud computing to rely on operational budgets (OPEX) are not capital expenditure (capex) budgets. Cloud computing provides faster time to market, chargeback capabilities, and the ability to enable and disable virtual machines (VMs) to meet fluctuating needs. Cloud is a very scalable environment, full of potential. However, speed, scalability, and virtualization do not - by itself - guarantee return on investment (ROI). All benefits of cloud remain potential and not actual if the cloud is not managed for maximum efficiency and cost effectiveness. It brings companies in the cloud or who are considering moving to the cloud to the very important thing: To achieve a solid ROI in cloud transformation, companies must be involved in structured and strategic cloud-computing capacity planning. Businesses can assess the actual ROI from:
 * Eliminates out of date hardware
 * Replace software that is no longer supported by vendors
 * Combining applications in a shared environment
 * Stop obsolete applications
 * Refreshing technology on an opex basis
 * Utilizing new technology, software, or applications

Eight key factors that can be measured must be considered to produce a comprehensive cloud-computing capacity planning worksheet.

1. Service Level Agreement
Service Level Agreement, or SLA, is a good place to start analysis because - based on its nature - can be quantified. Companies must establish SLAs for areas such as availability, storage, backup and recovery. All applications must be reviewed, and the appropriate parameters are set. The challenge when analyzing SLA is that application owners, business functions, and IT may have very different expectations based on considerations such as perceived criticality, complexity, and costs. Working independently from each other, these various fields can assume an SLA that is contradictory or a priority level for the same thing, which results in misunderstanding, conflict, and potentially increases business risk. It is important for all stakeholders to reach consensus for each assigned SLA, based on a thorough review of the needs and factors involved. Allocation of SLA that needs to be evaluated


 * Availability?
 * Storage?
 * Backup?
 * Recovery?

2. Usage Pattern
The goal of virtualization and cloud is to build infrastructure that maximizes utilization and capacity. Therefore, the study of application utilization patterns is very important.

For example, in the financial world, certain applications will run at full capacity during business hours. But when the business closes, the use of the application has declined. Therefore IT must try to use the same infrastructure to run night applications, maybe batch jobs or fixed income applications. Doubling in this strategy mode removes the resources (and costs) of the two idle systems 50% of the time by allowing one system to function at optimum capacity 100% of the time. For example in other situations, IT can determine that certain servers or VMs are consistently running at only 30% of capacity. With 70% of the system stilling regularly, there are great opportunities for restructuring and transformation. The goal of virtualization and cloud is to build infrastructure that maximizes utilization and capacity.

Planning considerations:
 * What is the usage pattern per application?
 * Where is the capacity not used?
 * What applications can be combined on a virtual machine for optimal utilization?

3. Workload Analysis
The utilization pattern reflects the current situation, workload analysis explores the potential impact of actions on the surrounding area - before the action occurs.

For example, a business might create a workload scenario to identify consequences when introducing a new product, service, or technology. They might estimate the effect of new market penetration. Companies that are very seasonal can estimate the difference between utilization in the slow months vs. months of heavy volume. This scenario helps determine trends and determine the additional capacity needed to handle the increase in workload.

Because of their close interaction, the pattern of utilization and workload analysis must be analyzed together with each other.

Planning consideration How the workload scenario changes because:
 * New products or services offered?
 * New technology is being utilized?
 * New market penetration?
 * Seasonal fluctuations?

4. The Process Excepted
Not everything can or must be moved to the cloud. Therefore, companies must establish a well-defined exception process. The exception process will ask questions such as:


 * Is this software supported on a virtual platform?
 * Is the required agility and scalability guaranteed in the proposed cloud environment, or are there potential limitations that can affect the performance of the application provided?
 * Is there foot-print memory that this application needs that cannot be handled in a cloud-computing environment?
 * Will security, compliance and regulatory requirements be met if the application and related data are transferred to the cloud?
 * Does the application integrate with other applications, and if so, will the cloud support this integration?
 * What availability is needed for this application, and will the cloud support the required SLA?
 * If a company outsource to the cloud, are there concerns with creating external dependencies for certain applications?

Companies must have procedures to deal with these exceptions to support and maintain a solid overall ROI.

Planning consideration Check each application by considering the following to determine eligibility to migrate to the cloud:
 * Support software
 * Availability
 * Agility
 * Scalability
 * Integration with other systems
 * Security, compliance and regulatory requirements
 * External dependence

5. Data Management
Before the data is moved to the cloud, it is better that policies that discuss data creation, access, retention, archiving and deletion must be established to build and maintain an optimized environment. Data may need to be reset, transferred to low-cost storage, or completely removed from the system periodically. Without a data management policy, it is easy to end up with lots of files that contain the same data, and keep all data forever. This "packrat" mentality consumes more server space from time to time, adding to the increasing costs.

What must be taken into consideration in planning it is:
 * Creation of data
 * Data access
 * Data retention
 * Archiving data
 * Data deletion

6.
configuration Cloud Computing, like almost everything else, can be customized. But customization always costs. Before considering adjustments, businesses must check the standard configurations available with the intended cloud environment and determine whether the pre-determined configuration will be suitable for applications and data migrated to the cloud. If so, it will save time, effort, complexity, and costs both in cloud migration and cloud maintenance.

Planning consideration Are predefined standard configurations available? Does the configuration match applications and data migrated to the cloud? If not, why not

7.
Recovery Plan Cloud Computing provides the potential to reduce both recovery time objectives (RTO) and recovery point goals (RPO). When planning to move to the cloud, RTOs and RPOs must be properly defined and analyzed with respect to applications and data migrated to ensure optimization of costs and business continuity. Planning considerations:
 * Purpose of Recovery Time Objectives
 * Destination Recovery Points (Recovery Point Objectives)

8. Business Requirements
One factor goes through all the points previously covered: that is, IT always has the responsibility to turn business requirements into technological solutions. For example, the sales department can indicate that they need to process 30 transactions per minute. IT must take business requirements and break down the impact on the computing cycle, memory footprint, input / output (IO), drives, etc. Based on the detailed information, technology solutions can be designed and implemented.

To engage in effective cloud computing capacity planning, IT needs to sit at a table with all businesses, hear their needs in their language, translate those needs into IT terminology, and deliver on business. IT also needs to reverse the communication process many times; for example, they might need to explain technology-oriented disaster recovery policies to all businesses in the event that recipients can understand and appreciate. Only when a business works together as a whole can the cloud environment be truly optimized.

Planning considerations: Have all business partners reached understanding and agreement on questions about what, where, when, why, and how to use the cloud to meet their business requirements?

Reference

 * Bhatti, Sudhir. "VMware Cloud Management." VMware Cloud Management. N.p., n.d. Web. 11 Sept. 2013.
 * Raz, Shahar M. "6 Reasons to Remember Capacity Planning in Cloud Computing Environments." Raztscom. N.p., n.d. Web. 11 Sept. 2013.