In an increasingly volatile, uncertain, complex, and ambiguous (VUCA) world, businesses are looking to become more agile, and fast-track innovation for their consumers by modernising their Information Technology (IT) infrastructure. The hybrid multi-cloud, comprising on-premise data centres and the public clouds, is emerging as their go-to model as businesses want to balance security and reliability for their mission-critical workloads with agility and scalability.
Therefore, today, IT has become more and more dependent on the infrastructure as a service layer (IaaS), a system of cloud that delivers virtualized infrastructure resources (compute, network, and storage) over the internet. Here is a guide to help accelerate innovation at the infrastructure as a service layer.
Why innovation matters in the IaaS layer?
This IaaS layer is also known as the control plane, used for managing and orchestrating everything, be it data or applications on the cloud, or whether it resides on-premise or on the public cloud. The IaaS layer is foundational to the efficiency and efficacy of the entire IT. Its optimization is critical to connecting the products and service offerings of businesses to the consumers.
Further, the pace at which cloud adoption is growing needs to match the optimization needed for scale, performance, agility, flexibility, and resiliency at the IaaS layer. The lack of innovation at the IaaS layer will severely impact businesses and cloud service providers, leading to a heavy impact on business revenue and growth. In essence, in the absence of innovation at the IaaS layer, the purpose of the digital transformation agendas pursued by businesses would be in jeopardy.
Moreover, today’s cloud data centres need to scale rapidly, capturing growth opportunities which means the IaaS layer must scale for management of compute, network, and storage infrastructure provisioning. The same pace is also critical to add performance, agility, flexibility, and resiliency in the IaaS layer along with scale, which is orchestrating all the infrastructure pieces for the end-users in the cloud world. The challenge and complexity in IT concerning the cloud, be it private, public, hybrid or multi-cloud, is more prominent at the IaaS layer, given it is the significant aspect in any cloud.
Why is scaling crucial in the IaaS layer?
The IaaS layer is the place where all the resources needed for the cloud users are getting modelled and orchestrated. If it is unable to scale to the demand, we will see bad results for any cloud. All major public cloud providers, namely Amazon Web Services (AWS), IBM, Google Cloud Platform (GCP), and Microsoft (Azure) are heavily dependent on their IaaS layer which is sitting in their data centres. Top layer applications onprem cloud applications like VMware vRealize Orchestrator, IBM MultiCloud Manager, ServiceNow Orchestration, BMC TrueSight Orchestration, etc are also depending on the IaaS layer for all provisioning.
The IaaS layer should be able to handle multiple requests coming from these top cloud layers. For that, it will need built-in high-availability, load balancing, which cannot be served from a single server of IaaS layer control plane but a cluster of IaaS layer control plane servers working together for better scale, performance, and resiliency.
These clusters of servers work together for load balancing for better concurrency and API return time which is most important in the cloud stack. The amount of scale needed for this demand for managing physical and virtual resources is increasing and today, the IaaS layer must manage thousands of compute servers, network, and storage elements within the data centres.
Hence, the IaaS layer's cluster design must be flexible to support an additional number of resources by dynamically increasing the nodes in the cluster of IaaS layers control plane to achieve higher scaling on demand and resiliency.
How does flexibility play a vital role?
Given the variety of demands from cloud consumers, the IaaS layer needs to support multiple architectures at compute (x86, pp64, mainframes), different technologies and vendors at network and storage levels. The open source communities like OpenStack and others have been playing a crucial role in making this possible under one IaaS layer and focusing on clustering of control plane nodes.
To summarise, innovations must continue at an accelerated pace in the IaaS layer to help cater to the evolving scenario. We are seeing the necessary focus from the industry, be it system integrators or operating system and IT infrastructure vendors or open source communities in this space. With these innovations at the infrastructure-as-a-service layer, cloud computing led transformation plans of businesses would see success in cloud adoption.
Akhtar Ali is Vice President, India Systems Development Lab and Technical Computing Development at IBM India.