Advanced Computing in the Age of AI|Thursday, August 13, 2020
  • Subscribe to EnterpriseAI Weekly Updates:  Subscribe by email

How to Balance the Multimodal IT Mix 

via Shutterstock

As organizations around the world transform their enterprise systems to embrace modern and agile technologies, multiple infrastructures for different workloads and applications are needed. Enter multimodal IT, a co-existence of traditional infrastructure, software-defined infrastructure and application-oriented architectures.

The term “mode,” in the context of IT, simply implies a type of IT infrastructure and associated set of processes. Gartner uses the bimodal concept to illustrate the existence of two types of IT – Mode 1 and Mode 2. Mode 1 IT typically implies traditional IT infrastructure (waterfall or ITIL processes and long-cycle times). Mode 2 implies software-defined infrastructure and agile technology (rapid development for digital innovation).

An IT migration is not usually a linear path from one to two. Multimodal IT suggests there are variations, combinations and two-way interoperability across a variety of technologies. The challenge for many IT managers becomes how to implement a plan that focuses not just on two modes but one that considers a holistic approach to the organization’s end-to-end operating model.

Building the Right Bridge

Identifying the best way to navigate a variety of combinations is crucial to a smooth implementation process. By building bridges across different technologies, IT can modernize traditional infrastructure and, at the same time, protect its investment by avoiding the disruptive approach of rip-and-replace.

The following scenarios of multimodal IT demonstrate how transitioning between the two modes is beneficial and complementary of each mode’s key strength:

Access to a Multitude of Deployment Scenarios:

An IT team uses a traditional IT infrastructure for security and reliability. The applications here are not required to be moved around. Another team is supporting a customer facing e-commerce application that is continuously updated with features and bug fixes. The workloads are deployed on a cloud to support scalability and flexibility. The team also uses a virtualized infrastructure to run its containers while another team uses the cloud for analytics and large data sets. Supporting workloads across IT infrastructures offers a variety of deployment scenarios that co-exist across traditional and software-defined infrastructure.

Multi-infrastructure Workload Mobility

A development team starts with developing container workloads on a traditional on-premises enterprise server that runs a container engine on a physical or virtual server (Mode 1). As the number of containers increases throughout the project, so will the requirement for orchestration in order to scale. In this case, the workloads may move to a software-defined infrastructure (Mode 2) where the compute, network and storage can be easily provisioned and deployed. The team can move workloads transparently from on-premises servers to the cloud and vice versa for testing and production. The development team can also benefit by using traditional servers and a software-defined infrastructure to maximize its development efficiency.

Blending a Variety of IT infrastructures:

If you are running databases (SQL, Oracle, SAP, etc.), it is likely these are running on a traditional infrastructure (Mode 1). However, many organizations are now transitioning to container applications to run business analytics. The containers running the microservices applications are on the software-defined infrastructure while accessing the back-end databases over traditional infrastructure. The IT team, in this case, gets the reliability and security of the traditional infrastructure while leveraging the benefits of containers for customer-facing value-add applications. This mixture of IT infrastructure enables the organization to run business applications on software-defined infrastructure while also accessing the back-end databases housed in the traditional infrastructure.

From a business perspective, recognizing the advantages of balanced multimodal IT can lead to maximizing the value for each mode and consequently enhancing the overall value of the IT organization.  From an infrastructure perspective, it’s important to have a system in place that allows for the necessary degree of interoperability and flexibility. The scenarios discussed above drive significant requirements for the underlying systems platform. Specifically, a multimodal operating system (OS) is required that can provide the foundation so traditional applications keep running, software-defined components are built seamlessly and application-oriented architectures are supported. The multimodal OS bridges the traditional and software-defined infrastructure and helps break the silos.

Overall, the vast majority of IT environments are fluid, complex and varied. Instead of envisioning a transition as a start/finish process, embrace the dynamic that comes from a variety of technologies that holistically deliver optimum business value when managed effectively. With the proper platform, tools and techniques, IT can deliver predictability, quality and an ability to neatly handle changing technology.

Raj Meel is global product manager at SUSE.

Add a Comment

Do NOT follow this link or you will be banned from the site!
Share This