Extended Service Life of Data Center Equipment – The Iceberg Below The Water Line For Service Providers
Today’s data center realities are driving a renewed focus on leveraging service providers to augment or even replace in-house infrastructure. Data center consolidation, assisted by the increasing density of systems, is a very common use case in which a service provider provides either a temporary or permanent landing for consolidated workloads. Public cloud adoption is also accelerating, forcing decisions regarding workload location, migration, and rationalization. According to a 2020 Gartner research report, “By 2025, 85% of infrastructure strategies will integrate on-premises, colocation, cloud and edge delivery options, compared with 20% in 2020.”
Background
Having spent many years managing service provider data centers – first at Datapipe, then at Rackspace – I’m keenly aware of the challenges of maintaining storage, server, and networking hardware, profitably, for extended periods. Though each type of equipment is unique in terms of its reliability and serviceability, general principles remain for building a multi-year strategy that balances profit and SLA delivery.
Enterprise service providers come in various shapes and sizes. Some provide their own unique infrastructure-as-a-service (IaaS), with hardware stacks designed to support a wide range of customer workloads. These IaaS layers tend to be relatively homogeneous, with minor variations based on single-vs-multi-tenancy, performance requirements, and capacity. Like any capital project, the goal of the service provider is to maximize the occupancy of that infrastructure for as long as possible, without having to make major changes. From a hardware standpoint, that means operating well past the normal warranty period.
Still others provide co-location services, hosting customer-owned equipment within managed racks and/or cages. SLAs will vary by service provider type and contract, but the nuances of systems maintenance, especially for mature assets, are an important point of negotiation.
By the Numbers
Consider the basic economics of a service provider delivering IaaS to their customers. Nearly a third of the cost is typically hardware, with nearly another 10% going towards maintenance. The largest cost item is usually labor, seen in the model below at almost 40% of overall costs to manage the environment.
So, it is vitally important that a service provider can affect the cost of the hardware – and maintenance – of their estate, particularly as that equipment reaches the out years of its useful life. Those years five plus can be highly profitable and potentially fund the eventual refresh of systems. Getting smart about the support of the environment can also play a role in reducing the labor cost component, the single largest line item.
Figure 1. Sample cost model for service provider
Data Center Systems Support Realities
It’s important to remember that the majority of system updates occur within the original warranty period – either three or five years, depending on system type and vendor. Firmware updates, service packs, security patches, and other routine maintenance items are most frequent in the first two years after release, and then tail off considerably in year three. Updates are infrequent after that period, and nearly non-existent in years six and beyond.
Figure 2. Example Mid-Range Array Firmwareu Lifecycle
What does that mean for contracting maintenance with the vendor? Most enterprise OEMs have a menu of maintenance options, from premium – and therefore higher-cost – options that deliver white-glove service, to more cost-effective support options that are well suited to more mature systems, including end of service life systems. Many OEMs actually call these services Multi-Vendor Support or Multi-Vendor Services (MVS). By contracting with OEMs to provide these MVS options, the customer gets the peace of mind of OEM-quality support, including access to firmware and other systems updates, but at a cost that is more aligned with the systems age and criticality.
While MVS is one of the lesser-known services offered by OEMs, each of the most recognized OEMs do provide MVS and have matured the offering significantly over the last several years. The ability of a service provider to maintain continuous OEM support rather than a third-party maintainer (TPM) provides a greater level of comfort to both the service provider and their end customers, along with tangible risk reduction.
Call to Action
The role of the data center service provider has never been as important to enterprise IT as it is today. Modernization initiatives are putting pressure on everything from infrastructure and talent acquisition, to application time to market. Service providers offer a range of options that can help extend the capabilities of in-house IT. By getting smart about extended support of mature storage, server, and network systems in their data center, service providers can manage and improve the profitability of their services without sacrificing quality and SLA performance.