Thoroughly Modern: A Practical Primer For The IBM i Cloud Journey

March 11, 2024 Lief Morin

Cloud has evolved significantly from its roots in the shared services era of IBM mainframes in the 1960s and 1970s. While some parallels can be drawn, today’s cloud landscape emphasizes flexible capacity and pricing models over traditional outright purchases of computing resources like the IBM i platform. The prevailing trend towards major cloud platforms such as Amazon Web Services, Microsoft Azure, Google Cloud, and IBM Cloud has underscored the importance of a shared architecture approach for cost-effectiveness and business prosperity.

Most businesses rely on heavily modified solutions for their core business applications, so they find significant benefits in a shared cloud solution that provides the infrastructure, security, and other risk-mitigation qualities needed to ensure continuous operations. Those benefits are now fully realized in the IBM i cloud marketplace and with the long-term commitments made by IBM and their cloud partners, the time to move to the cloud has never been better.

From the outset, the genius of the IBM i platform is that it is a completely integrated system, that included the OS, database, programming languages, application development tools, and security needed to deploy mission-critical applications that run a business, all in the box. This stood in stark contrast with other platforms; IBM i is, and has always been, a Platform as a Service architecture. This has made the IBM i system attractive as it enables organizations to focus on running applications that drive the business.

Thirty years on, the cloud has changed the nature of technology consumption forever, and in turn, applications, business models, and entire industries. With robust cloud offerings, the IBM i has changed with it.

Deciphering Cloud Architectures

There are many cloud architectures, and even more cloud providers. That trend is not likely to slow down, and the increase in complexity has grown exponentially. For the purposes of this article, we will focus on the three that are most consumed in the IBM i community: private, public, and hybrid clouds. A good place to start is with the authoritative definition of a cloud as provided by NIST (National Institute of Standards and Technology). It describes these five key attributes:

On-Demand Self-Service: Users can provision compute, network, and storage as needed, automatically, without human intervention from the cloud provider.
Broad Network Access: Cloud capabilities are available over the network and using standard clients.
Resource Pooling: The cloud pools resources and sells them to multiple tenants, with resources dynamically assigned and reassigned based on demand. Users cannot specify where their capacity is specifically located, but can specify a datacenter region or country.
Rapid Elasticity: Capacity can be scaled up rapidly and scaled back to zero as needed, returning capacity to the pool.
Measured Service: Clouds automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service – such as storage, processing, bandwidth, and active user accounts. Resource usage can be monitored, controlled, and reported, providing transparency for both the cloud and the user.

The cloud definition can be more flexible than NIST’s original version, so long as key characteristics are preserved. For example, IBM i cloud providers manage provisioning controls for application stability. Public clouds, known for their vast capacity, initially hosted temporary workloads but are now crucial for various industries’ key applications. IBM i platforms run stateful ERP and custom apps, making them vital for organizations. Private clouds mirror public clouds but lack shared tenancy, except for specific cases. Hybrid clouds blend public and private architectures, like DRaaS for replicating platforms to a cloud partner in the IBM i community.

Why Are Businesses Moving To Cloud?

Many IBM i clients choose to pursue a cloud strategy to increase service, reliability, availability, and security. Others evaluate cloud services as a method of reducing costs. Sometimes these elements line up based on the current environment’s profile being evaluated, but often there are benefits that extend well beyond these qualities and economics.

A cloud solution can provide the intangible benefits associated with some of the five qualities noted above. Elasticity, resource pooling, and autonomic optimizations, are primary amongst them. At its core, the ability to consume only what is needed, and then add or remove resources to a cloud instance in real time is always better than investing in dedicated and deliberately over-provisioned equipment.

Navigating The Landscape Of IBM i Cloud Providers

After many years of development, there are now credible, mature IBM i cloud providers worldwide. Some examples of a mature IBM i cloud provider are as follows:

Hosted in a carrier-grade datacenter
Employ dedicated network engineers whose mission is the security and stability of those environments
Designed and deployed by veteran IBM i architects
Managed by a team of skilled talent, available 24x7x365
Tested and certified by external audit and compliance organizations

This market is much smaller than the market for X86 cloud infrastructure though. Many providers are small or are an amalgamation of those same small operators creating an unacceptable risk profile. Others are exclusively SaaS providers, and some have ties into hyperscaler clouds; even IBM has its own offering.

However, most come with limitations. Some do not offer managed services, which defeats the goal of freeing up valuable internal resources to concentrate on running the business, and enabling the organization to hedge against resources who may be retiring. Further, most IBM i cloud providers are not flexible in adapting to an organization’s unique architectural requirements. Fewer still can provide Intel infrastructure with low latency interconnectivity to transfer large volumes of data amongst those environments.

Every mature IBM i provider should be able to provide these basic services:

Infrastructure as a Service
Backup and Recovery as a Service
Disaster Recovery as a Service
Managed Services (remote, as well)

All should be battle tested in handling real DR events and all should be able to assist in the creation of a technical disaster recovery plan.

Some go further than the above, including the ability to support old operating systems like V5R4, co-locating client equipment (such as an old power system whose application cannot be moved), and acting as a Backup and DR target for on-premises replication solution.

If you have limited resources and cannot have your programmers also managing the stability and security of your system, best-in-class providers also offer custom managed support that may include line-of-business support.

The Migration Process: Factors To Weigh In

If you are considering moving IBM i workloads to the cloud, you can just lift and shift your applications from your datacenter to that of a service provider of your choice. It is, however, never that easy. To start, the move to cloud requires a commitment from leadership and shift in culture to ensure a successful transition. Only then can the technical elements be defined and a plan to move put in place.

There are only two viable solutions – logical replication, or the enduring strategy of restoring from a tape backup. Not all cloud providers provide both, evaluate a potential partner on their abilities to support both approaches.

Each approach has its pros and cons. Logical replication provides the ability to replicate the data in real time and execute a cutover with extremely limited downtime. It is, however, more expensive if a replication tool is not already in place. A tape restore is a tried-and-true method but requires major downtime at the cutover point. Both approaches can be tested well in advance of a cutover to ensure all connectivity, applications, and connectivity are in place. Select an IBM i provider that can provide demonstrable proof of completing migrations using both solutions.

The single biggest migration hurdle to overcome, however, are third-party license fees that may be charged for serial number transfers, and some are significant. Code that is developed in-house does not have this limitation. This challenge is unique to the IBM i ecosystem and occasionally represents an insurmountable barrier for many to overcome. It is vital to take an inventory of your applications and contact your software providers as part of an economic evaluation of a cloud solution.

IBM has made strides towards reducing the impact of these with the implementation of Virtual Serial Numbers, however, many third-party organizations do not participate and are further adapting their licensing model to fingerprint an operating system environment in lieu of being bound to a specific serial number.

As an unfortunate by-product, virtual serial numbers dramatically limit the ability to implement local high availability using PowerHA and hinders Disaster Recovery testing and actual recovery events as those license keys are needed in advance of those activities occurring.

Select a cloud partner who has been through countless migrations, this is crucial in ensuring a well-architected and successful solution.

When Is The Right Time To Move?

Focus your key timing inquiries on your business model. Questions to consider:

When is your budget finalized?
When is your peak season?
Are there anticipated resource constraints that might accelerate a move?
Is there an identified business, application, or system risk that is unacceptable?
What other projects and priorities are part of the technology roadmap for the organization?

The performance gains of the recently delivered Power10 architecture, new IBM flash storage arrays, and powerful monitoring software have enabled cloud providers to amplify the benefits you gain. Moreover, IBM’s long-term roadmap reaffirms that your applications will keep running smoothly, including mission-critical ones, in any consumption model, especially in the cloud. The subscription pricing also aims to facilitate a seamless shift from on-premises to a cloud provider.

As with every technology, the march of time will always bring better performance and enhanced capabilities. So, there is never a bad time to evaluate a cloud option.

Wrapping Up: The Implications And Advantages Of Cloud Migration

Transitioning IBM i applications to the cloud offers you a great chance to upgrade service levels and boost organizational efficiency. Engaging with the right cloud partner after a careful evaluation is crucial for a successful transition. By understanding cloud architectures, identifying migration catalysts, assessing migration factors, and timing the transition well, you pave the way for a smooth and effective migration process that aligns with your business goals.

If you are considering a move and would like to explore your options, shoot me an email, I would be happy to chat.

Lief Morin is the general manager of cloud at Fresche Solutions. He started his career in 1990 as a programmer and field engineer on the AS/400 and RS/6000. He founded and ran an IBM solutions provider for 20 years and joined Fresche after selling that business. He specializes in cloud services for IBM i.

This content is sponsored by Fresche Solutions.

This Issue Sponsored By

Table of Contents

Content archive

Recent Posts

Subscribe

Pages

Search