In a previous post, we spoke about about how some companies can find themselves accidentally locked into difficult situations when they adopt multi-cloud applications without an overall strategy. We also discussed the need to define critical components of a multi-cloud strategy. So, what does that look like? Here are six ways you can start preparing for hybrid cloud.
1). Define your Goals.
Ensure your team doesn’t purchase solutions in search of a problem. When looking for cloud technologies, first define the goals that you are looking to achieve. Is it reduced cost? Improved efficiency? Ease of management? Reduced complexity? By defining goals up front, that will guide the team’s choices during the process of evaluating technologies.
2). What Problems Are You Solving For?
Typically, there are several problems that the IT team is challenged to address. The first step is ensuring that you prioritize the most mission-critical problems. This will guide the team in selecting a solution that drives the most business value. Enclosed are some questions to consider:
– Are there increasing infrastructure costs due to datacenter silos”?
– Is ease of management an issue?
– How about datacenter space and power issues?
– Is reliability an issue and is high availability a requirement?
– VDI or database performance Issues?
– How about security & governance requirements?
– Growing problems with storage efficiency?
– Backup & disaster recovery issues?
– Validation and certification of 3rd-party software applications?
3). Implement an Intelligent Hybrid Cloud Policy Engine
Hybrid cloud environments lend themselves to a world of competing requests for data, bandwidth, and processing resources. Users need to ensure that their most critical applications reserve the most compute power and the lowest-latency connections, while also ensuring that their most sensitive data is secure. In all but the simplest environments, this is a job that many times, is too complex to be performed by human administrators at scale.
Quality of Service (QOS) or Intelligent Policy Engines can do this job in both on-premise and multi-cloud environments, however it’s important to select platforms with this capability very carefully.
Many vendors sell their platform in the form of a “black box”—that is to say, that policies aren’t configurable by either the administrator or the end user. Without the ability to configure the policy engine for constantly changing demands, companies will find that some solutions will have difficulty managing multiple application requirements on a single platform….and will continue to drive the building of datacenter silos.
Intelligent hybrid cloud solutions should enable allocation of CPU, Memory, Network, and Storage resources, on the fly to guarantee application performance based on the needs of the end-user. Enclosed are some questions to consider when evaluating competing technologies:
4). Map Data Storage Efficiently
As companies prepare a move of key applications to cloud environments, many times there is a huge amount of associated data to contend with. Prior to the move, they will need to understand the best data reduction, protection, and migration strategy. This will prevent accidental loss of data, ensure security, and allow for efficient data migration. There are well known technologies that enable this:
Deduplication – Finds duplicate data and ensures that only one unique instance of that data is retained to save storage space. This reduces the amount of data being migrated to the cloud. Deduplication technology is compute intensive and does not work well on video data. This consideration should be explored to decide if best is for primary or secondary storage use cases.
Compression – Is the process of modifying, encoding, or converting the structure of data at a bit level in such a way that it consumes less space on disk. This technology many times is leveraged in conjunction with deduplication to reduce storage required.
Erasure Coding – Is a method of data protection in which data is broken into fragments, encoded, and pieces of the data are stored across different locations in a storage array or cluster. This technology ensures that no data is lost, in the event of a multi drive or a node failure and is superior to traditional RAID and “Data Replication” techniques implemented by many hyperconverged storage vendors.
Encryption – Is the process of converting information or encoding data, to prevent unauthorized access. “Data at Rest” and “Data in Flight” encryption technologies should be considered, depending on the sensitivity of the data being migrated to/from the cloud. Security and compliance requirements are key aspects of any cloud implementation.
5). Ensure Flexibility Between Cloud and On-Prem Environments
Implementing a Hybrid Cloud environment means balancing between Public Cloud and On-Premise hosted applications and data. Ideally, companies should identify solutions that work well in both environments and have the ability to migrate to and from these environments easily. This can be done based on the environment that provides the most business value. When companies find a solution that works well for them on premise, however isn’t offered in the cloud, they should consider the implications to their long-term strategy and flexibility.
6). Business Continuance, DevOps, & Cloud Flexibility
Part of the hybrid cloud appeal is that some companies could conceivably run their entire infrastructure in the public cloud. For example, during unexpected downtime, companies can spin up and spin down cloud instances of mission-critical applications and data, quickly and efficiently. This can also be done when development projects require infrastructure for a relatively short period of time. Clone your applications and data, replicate the clone to a public cloud and quickly spin up a development environment. The key benefit is that you pay for only the resources that have been consumed. This capability also enables movement between different public cloud vendors to take advantage of new capabilities or cost. Solutions that enable these capabilities deliver on business continuity, future DevOps requirements, and cloud flexibility.
Build a Multi-Cloud Strategy with Pivot3
Pivot3 offers technology to seamlessly unify applications and data between multiple clouds, developing connections between different public clouds environments, SaaS applications, and allowing administrators to transfer workloads seamlessly between public cloud and on-premises cloud environments.
Hybrid cloud is the future of the enterprise. For more information, download our Pivot3 Cloud Edition Solution Brief. See Pivot3’s intelligent hyperconverged platform for hybrid cloud and IoT solutions in action via a Pivot3 Weekly Webinar