Organizations require new products to manage their cloud systems in a coherent fashion and VMware continues to extend vSphere to serve multi-cloud needs. Still, to take full advantage of a multi-cloud system, organizations must implement a few best practices to ensure their systems don't negatively affect each other.
Organizations continue to migrate their workloads to the cloud, but its design is much different from legacy systems. VMware can help organizations maintain consistent infrastructure and operations across clouds with products such as VMware Cloud Foundation and CloudHealth. These can help improve visibility, cost management, security and governance.
Still, organizations must slowly introduce new cloud systems to staff, identify workloads that benefit from a public cloud migration and establish a long-term systems management plan to maximize vSphere's cloud potential and improve their VMware multi-cloud strategy.
The significance of the multi-cloud market
There's no doubt organizations are embracing the cloud. The worldwide public cloud services market grew from $185.2 billion in 2018 to $233.4 billion -- a 26% year-over-year increase, according to IDC. As a result, organizations have a wide and growing mix of on premises and the cloud that often introduce management challenges.
"Multi-cloud architectures are introducing a new wave of management complexity as developers and business groups implement cloud services and tools that best align with their application and business innovation road maps with limited regard for corporate preferences. The introduction of containers, microservices and Kubernetes creates further complexity," according to Mary Johnston Turner, research vice president for future of digital infrastructure agenda at IDC.
In the past, organizations often relied on legacy data center management tools.
"VMware has been trying to bring vSphere's multi-cloud management capabilities up to par with what it has offered for legacy systems," said Torsten Volk, managing research director of hybrid cloud, software defined infrastructure and machine learning at Enterprise Management Associates.
Legacy systems have a different design compared to modern systems. In the past, information flowed in set patterns, which made it easy to observe and understand data. These types of management tools observe data at points along a connection, collect information as it moves from place to place, and compare performance metrics to real-world scenarios.
Still, this approach doesn't work with all cloud applications because some rely on microservices. Microservices split applications into separate services, which then run their own unique processes and manage their own databases. Information also flows in a less linear fashion.
As a result, admins should have a chance to become comfortable with new cloud systems. Organizations shouldn't overwhelm them and introduce all changes in one fell swoop. A better strategy is to have admins focus a single area of expertise, such as network response time monitoring.
IT teams must illustrate the differences between the old and the new; let them experiment with these new systems until they feel comfortable. Then move them to another management area.
Identify workloads suited for the public cloud
Corporate applications' complexity level, maintenance requirements and effect on a system aren't always equal. Some applications offer a low risk, high reward model when organizations move to the public cloud.
"When embracing public cloud, businesses often start with storage use cases: disaster recovery, backup and data replication," Volk said.
These applications are often a storage infrastructure purchase and can port fairly easily from a data center to the cloud.
A workload, such as a sales system, has many pathways that connect to a variety of data sources and provides information used by other applications. The migration of such an application often requires tedious rewriting and retooling of the system, which increases the time, manpower and resources required to ensure successful migration.
Cross-train support staff
In many cases, multi-cloud deployments span both on-premises data centers and the public cloud. Organizations might keep a VMware application on premises because the cost to rewrite and migrate it to the cloud is high. Greenfield applications are often good candidates for the public cloud because they don't have the technical debt that often comes with legacy applications.
Understanding how both legacy and greenfield applications work with the cloud becomes key to long-term management success. Typically, organizations dedicate personnel to specific applications to develop deep expertise with a tool or workflow. Regardless, all applications have an effect on system performance.
In some cases, noisy neighbor applications can consume excess resources and affect the performance of an on-premises system. Cross training employees so they gain an understanding of how other systems function can improve troubleshooting operations.
Adopt a long-term view
It's taken some organizations years or even decades to build up their current VMware infrastructure. The public cloud offers positive changes in how organizations deploy systems.
"VMware has done a lot of innovative work in areas, like security, that in the long term will benefit organizations," Volk said.
In the short term, the switch to the cloud adds complexity and cost to VMware support, so organizations must consciously trade more immediate shortcomings for long-term improvements.
The mix of on-premises and cloud applications continue to change as the public cloud gains traction. Consequently, organizations must devise a new approach to manage their infrastructure. Many organizations use vSphere to ensure their systems perform well, and multi-cloud strategies are emerging to help them with their transition to the cloud.