For the past few years, cloud computing has been heralded as the next biggest trend in enterprise IT with benefits set to meet every business requirement.
Market analysts from leading agencies such as Gartner and IDC have in their independent opinion estimated an exponential growth of the cloud market with predictions estimating it to reach a value of $72.9bn by 2015. Already organizations across the Middle East are approaching cloud solutions vendors with the hope of deploying a model which will dramatically cut operational costs which still acting as a vehicle to drive business.
And while such discussions are well underway, there still is a fair level of confusion regarding the actual path to the cloud.
George DeBono, General Manager, Middle East & Africa at Red Hat says that one of the reasons behind this bewilderment is the sheer number of offerings available under the broad cloud umbrella. The wide array of cloud strategies and enabling technologies that have been made available in the past 12 to 36 months offer varying levels of performance in factors such as access, dynamic allocation, scalability, application hosting and management. It is no wonder then that IT managers are completely overwhelmed by the cloud clutter.
To achieve effective utilization and cut the cost of procuring and managing physical servers, virtualization has become the norm in large enterprises across the Middle East.
Virtualization-related cost benefits, however, don't directly address the business need for more agile IT. Cloud computing on the other hand does this and it is time to transition from virtualization to the cloud.
IT managers however need to be wary of hastily migrating to the cloud without a well thought up action plan because they risk needless expenditure on technologies which do not meet the unique requirements of the organization. Or, the organization may choose the right solution but fail to utilize it to the maximum potential because of an improper deployment.
By following the guidelines described below, IT managers can get their private cloud deployments on track right from the start. The activities below mark out the key steps involved in migrating from the enterprises' existing virtualized environment to the private or hybrid cloud.
And while such discussions are well underway, there still is a fair level of confusion regarding the actual path to the cloud.
George DeBono, General Manager, Middle East & Africa at Red Hat says that one of the reasons behind this bewilderment is the sheer number of offerings available under the broad cloud umbrella. The wide array of cloud strategies and enabling technologies that have been made available in the past 12 to 36 months offer varying levels of performance in factors such as access, dynamic allocation, scalability, application hosting and management. It is no wonder then that IT managers are completely overwhelmed by the cloud clutter.
To achieve effective utilization and cut the cost of procuring and managing physical servers, virtualization has become the norm in large enterprises across the Middle East.
Virtualization-related cost benefits, however, don't directly address the business need for more agile IT. Cloud computing on the other hand does this and it is time to transition from virtualization to the cloud.
IT managers however need to be wary of hastily migrating to the cloud without a well thought up action plan because they risk needless expenditure on technologies which do not meet the unique requirements of the organization. Or, the organization may choose the right solution but fail to utilize it to the maximum potential because of an improper deployment.
By following the guidelines described below, IT managers can get their private cloud deployments on track right from the start. The activities below mark out the key steps involved in migrating from the enterprises' existing virtualized environment to the private or hybrid cloud.
Establishing Goals Based on Business Objectives
Building a private cloud designed specifically for an enterprise must start with a business discussion. This should include a mix of key stakeholders- typically a team with developer, users, and, more importantly, externally facing product, marketing, and sales managers. The discussion should revolve around how the cloud can accelerate business processes or transform business offerings.
Business objectives, constraints and the expected benefits all need to be highlighted during this phase. The vital questions which need to be addressed are: What gives the business its competitive advantage? Where does technology drive opportunity, but is limited by having IT in the critical path? What new things can IT do to help the business? What limitations does the current IT situation impose? How will the enterprise benefit if the deployment is a success? What KPIs will be used to measure cloud success?
If businesses decide for a path to cloud and an architecture for cloud on a company level, they should be taking into account that this will be a decision to last for 5-10 years.
Therefore they should take into account to be able to be vendor neutral on their cloud architecture and their cloud management stack as much as possible, to not be locked-in into a single vendor, or economic model and not to be locked-out of future regulations or innovation opening new possibilities in cloud.
Business objectives, constraints and the expected benefits all need to be highlighted during this phase. The vital questions which need to be addressed are: What gives the business its competitive advantage? Where does technology drive opportunity, but is limited by having IT in the critical path? What new things can IT do to help the business? What limitations does the current IT situation impose? How will the enterprise benefit if the deployment is a success? What KPIs will be used to measure cloud success?
If businesses decide for a path to cloud and an architecture for cloud on a company level, they should be taking into account that this will be a decision to last for 5-10 years.
Therefore they should take into account to be able to be vendor neutral on their cloud architecture and their cloud management stack as much as possible, to not be locked-in into a single vendor, or economic model and not to be locked-out of future regulations or innovation opening new possibilities in cloud.
Analysis of Infrastructure
In the near future, enterprises are likely to have IT infrastructures which involve a mix of physical, virtual and cloud-based resources deployed at either internal or external locations. Instead of thinking about each environment separately, organizations need to adopt a portfolio view of all of their IT resources. This will help IT managers determine what workloads will be best deployed in which environment. Also it will facilitate to improve process efficiency gains over all infrastructure silos.
Because of factors such as cost, flexibility and scalability, quality of service, physical location and lifecycle phase, each environment will have its own share of strengths and weaknesses. The portfolio view helps gain a perspective of this paving the way for the next phase of planning.
Because of factors such as cost, flexibility and scalability, quality of service, physical location and lifecycle phase, each environment will have its own share of strengths and weaknesses. The portfolio view helps gain a perspective of this paving the way for the next phase of planning.
Identification and Allocation of Workloads
Assessing the current workloads creates a snapshot which helps IT managers identify which of them would be a good fit for the private or hybrid cloud environment. Organizations that have already opted for a high degree of virtualization would already have a detailed list of hardware and software assets that have been identified for consolidation. Coupled with review of disaster recovery plans which would include a list of business-critical applications, this will create a snapshot of the organization's existing workloads.
Each workload should then be prioritized according to attributes such as business objectives, resource requirements, usage stability, heterogeneity, data, security and compliance, critical dependencies and performance requirements. Based on this, a plan detailing which workloads will be migrated to the cloud environment can be formulated.
Each workload should then be prioritized according to attributes such as business objectives, resource requirements, usage stability, heterogeneity, data, security and compliance, critical dependencies and performance requirements. Based on this, a plan detailing which workloads will be migrated to the cloud environment can be formulated.
Evaluating Cloud Computing Models
A good first step towards cloud computing would be to adopt a hybrid model which utilizes resources already available in the organization. However each cloud model offers it share of advantages and disadvantages which make certain models better suited to certain organizations. Below are listed a number of considerations which should be evaluated when selecting the cloud model.
Service Quality- Not all cloud providers offer arrangements which guarantee a contractual Service Level Agreement (SLA) with penalties. If quality of service and uptime is vital, such an agreement is an absolute must.
Hybrid Model- If the scalability requirements are high and dynamic in nature, the organization may have to turn to a public cloud provider. If however peak in usage are predictable, the private cloud may still be a good option. It is to be expected though, that due to easier regulation and more community clouds coming up hybrid models will be more prevalent in the next few years.
Security and Compliance- For sectors such as banking, healthcare and government where security is of utmost importance, stringent limitations may rule out some cloud options. Risk tolerance and business criticality of workloads vary considerably among companies. Community clouds can be a good options for those enterprises.
Heterogeneous workloads- Although virtualization abstracts applications from underlying resources, some cloud solutions work only with specific virtualization technologies. Others are optimized for specific programming languages. These constraints may rule out some platforms that are limited to a specific technology.
Openness- Controlling the economics and deployment options for a cloud architecture is key to retain flexibility for the future. Infrastructure, Public Cloud and Virtualization Deployment options should be wide, extensible and governed by a community. By abstracting away from this, enterprises can become through utility providers to their internal and external customers.
Whatever be the cloud model ultimately deployed, it is advisable that in the initial stages low priority workloads are first tested in the newly set up environment in order to ensure that it meets the performance and quality of services specifications. The benefits of lower operations costs, improved business agility and greater scalability and manageability make a compelling case for pursuing a private or hybrid cloud strategy. Careful planning and evaluation are vital to success and by following the above guidelines, organizations can hit the ground running with a widescale cloud enterprise architecture, which is future-proof and open for the innovations to come.
Service Quality- Not all cloud providers offer arrangements which guarantee a contractual Service Level Agreement (SLA) with penalties. If quality of service and uptime is vital, such an agreement is an absolute must.
Hybrid Model- If the scalability requirements are high and dynamic in nature, the organization may have to turn to a public cloud provider. If however peak in usage are predictable, the private cloud may still be a good option. It is to be expected though, that due to easier regulation and more community clouds coming up hybrid models will be more prevalent in the next few years.
Security and Compliance- For sectors such as banking, healthcare and government where security is of utmost importance, stringent limitations may rule out some cloud options. Risk tolerance and business criticality of workloads vary considerably among companies. Community clouds can be a good options for those enterprises.
Heterogeneous workloads- Although virtualization abstracts applications from underlying resources, some cloud solutions work only with specific virtualization technologies. Others are optimized for specific programming languages. These constraints may rule out some platforms that are limited to a specific technology.
Openness- Controlling the economics and deployment options for a cloud architecture is key to retain flexibility for the future. Infrastructure, Public Cloud and Virtualization Deployment options should be wide, extensible and governed by a community. By abstracting away from this, enterprises can become through utility providers to their internal and external customers.
Whatever be the cloud model ultimately deployed, it is advisable that in the initial stages low priority workloads are first tested in the newly set up environment in order to ensure that it meets the performance and quality of services specifications. The benefits of lower operations costs, improved business agility and greater scalability and manageability make a compelling case for pursuing a private or hybrid cloud strategy. Careful planning and evaluation are vital to success and by following the above guidelines, organizations can hit the ground running with a widescale cloud enterprise architecture, which is future-proof and open for the innovations to come.
Thanks for sharing valuable information. Your blogs were helpful to AWS learners. I
ReplyDeleterequest to update the blog through step-by-step. Also, find the AWS news at
AWS Online Training