Skip to main content

Designing a Cloud Strategy Before a Cloud Migration

By September 19, 2018November 2nd, 2018No Comments5 min read

Thinking about jumping into the cloud. Prepare migration to the cloud with IT inventory and key figure reconciliation.

By 2025, cloud storage usage will be so firmly rooted in everyday business that there’s more to it than talking about the cloud. This was forecasted in a recent survey conducted by the American software provider Citrix. There is still a lot of talk. For example, whether a private, hybrid or multi-cloud is best. It depends on the industry environment, business activities, business goals, IT requirements and budget, and the IT know-how of the workforce. For example, it can improve the global distribution of a company if all reporting staff have a reporting tool at their disposal. In that case, it makes sense to run this workload in the cloud. You can also set up on-premise work,

Even for this example, the principle is: Before a cloud migration, a company should design a cloud strategy. Therefore, a business analysis has to first uncover where there is business potential and where core processes can be streamlined and digitized. Often the question arises as to whether an SAP application will continue to run better in its own data center or whether a leaner customer relationship management (CRM) from the cloud will suffice. Part of the answer comes from an IT inventory.

Clean up the stock and clear out

As a result of the inventory, a company knows when hardware and software are written off. The inventory also divides the application landscape into applications that can or must be moved 1: 1 to a public cloud. In a future cloud, operation remains to clarify what the company has to reprogram or buy. It is not possible to generalize beforehand what definitely belongs in the cloud. For example, an application stays in the data center for compliance or regulatory purposes. The General Data Protection Regulation (GDPR) requires companies to establish and document transparent processes for the handling of data. A company can not delegate responsibility for data security and protection to the provider.

In the next step, key figures help to decide pro or contra cloud. Through key metrics such as Recovery Point Objective (RPO) and Recovery Time Objective (RTO), a company determines how high-availability its application must be in the cloud. If there is absolutely no data loss, the provider must guarantee a zero RPO in Service Level Agreements (SLAs). With an agreed RTO of 60 seconds, this is exactly the time for data recovery.

Set other important time factors

In addition, a company should define the deployment time for itself. How fast a cloud structure reacts can be assessed with provisioning metrics. Information about user satisfaction exists in cloud operation when a company combines multiple metrics such as workload performance, deployment time, number of workloads, and mean time to failure (MTTF). This shows whether users achieve the expected results.

Whether cloud service providers meet their own performance requirements, it is next to check. Companies should first look at the service KPI (Key Performance Indicator) availability. If the availability of the SLAs is 99.9 percent, a company can not use the cloud for an average of 42 minutes per month. Cloud providers typically use a ticketing system to troubleshoot issues, changes, and help requests. It is recommended to set short response times of the cloud provider to requests in the SLAs.

Capacity and cost issues

A workload should be highly available and performant in the cloud, which requires a certain storage capacity. However, today’s storage space can be reduced by as much as 70 percent using snapshot technology, data compression, deduplication, and cloning. Nevertheless, the required performance and storage capacity belongs in the SLAs.

Ultimately, the costs decide which cloud migration pays off. For a planned SaaS such as Salesforce or Office 365, it’s easy to figure out the cost. These should be lower than those of the in-house service to be replaced. Otherwise, it’s a good idea to drill down into the monthly cloud operating costs by users and business groups. So later it’s easy to find deviations from the predicted cloud usage. A regular comparison of infrastructure costs caused by the local data center with cloud operating costs indicates rising or falling cloud acceptance.

Perspectively thinking about data management strategy

IT Inventory and Key Performance Metering deliver applications and services that run cost-effectively in the cloud. The result may be the prospect of efficient software and technology deployment, real-time data analysis, better enterprise-wide collaboration, agile app development, artificial intelligence-based customer service, or networked production based on machine learning. Prior to cloud migration businesses should consider that they need a solution for hybrid or multi-cloud environments that makes it easy to manage and protect their enterprise workload data. It also requires synchronization of data used on-premises and in the cloud, which can be realized through data management platforms.

Contact us today to learn about Bleuwire™   cloud services and solutions on how we can help your business.