Data center consolidation: Strategy and best practices
The modern pace of data creation is staggering. The average organization produces data constantly—perhaps even continuously—and soon it’s investing in servers to provide ample storage for that information. In time, and probably sooner than expected, the organization accrues more data and outgrows that server, so it invests in multiple servers. Or that company could tie into a data center, which is built to accommodate even larger warehouses of information.
But the creation of new data never slows for long. And if an organization takes its new metrics and performs extensive data analysis on them, one result will be that even more data is created from that analysis. At some point, the organization’s data center assets may outgrow even that storage and the company will need to use multiple data centers. Or different departments within that company might opt to use other data centers for certain informational workloads, even though such action ignores best practices.
Regardless of their origin, there are problems like server sprawl that are created by holding information in disparate physical locations. These are largely problems of inefficiency and require a solution that streamlines data storage processes and keeps an organization’s data resources safely organized within a logical framework. And that’s why companies need an effective data center consolidation strategy.
What is a data center?
Because the term applies to installations of various sizes, we should be clear about what constitutes a data center. First and foremost, a data center is an actual place, whether that be an entire building or a single room. It’s a physical space with definable characteristics that’s dedicated to housing and hosting IT infrastructure.
In their genesis, data centers tended to be singular creations developed for the sole use of a single company. Over time, the term came to encompass large, hyper-scale data centers run by cloud service providers, which host numerous organizations and individuals from different companies. For example, IBM operates over 60 IBM Cloud Data Centers in various locations around the world.
There are three types of data center facilities that vary according to their business uses and the workloads they handle:
- In enterprise (or on-premises) data centers, an organization hosts all of its data on-premises, usually because a maximum sense of security is desired.
- Meanwhile, public cloud data centers are hosted within cloud environments by large corporations and offer the convenience of being accessed through a basic Internet connection.
- Lastly, there are managed data centers and colocation facilities for those organizations not really equipped to run their own data centers but who wish to avoid using the shared resources found in a public cloud data center. In managed data centers, an organization leases the necessary hardware from a data center provider, which then administers the management of that data center. In colocated facilities, an organization owns the infrastructure and leases all of its hardware from a data center provider.
What is a data center consolidation strategy?
What data assets do you have and where will you keep them all? And is there a better way—one that offers greater efficiency? These are the guiding questions behind data center consolidation.
A data center consolidation strategy is a plan an organization creates and implements to shrink its data-storage processes and streamline its system of data management.
Usually, when we discuss data management services, we’re talking about how to implement them for the gathering and harvesting of data and how to scale them upward for ever-increasing amounts of data. But in the case of data center consolidation, we’re really advocating for the collapsing of certain systems and the integration of their data into a more centralized location. In other words, data center consolidation is primarily a reductive process—one in which the amount of a company’s data isn’t changed, but the number of its various storage systems typically is reduced.
How does a data center consolidation strategy work?
A strategy built around data center consolidation doesn’t need to be very complicated, but that consolidation plan will depend a good deal on other variables, like the number of different types of data an organization needs to retain.
At its core, data consolidation efforts are a fairly straightforward activity that start with gathering organizational data from various sources. Then, the data gets improved as needed by cleaning it and combing it for possible errors. Finally, the optimized data gets stored in one location, where it can be accessed most readily.
At least that’s the simple version. In reality, for most companies, crafting a data center consolidation strategy tends to be a much larger and more involved affair. There may be multiple data centers’ worth of data that needs to be gathered, cleaning that data may actually require several steps, and the organization must make important decisions about current data needs and smart projections about future data storage requirements.
Another key step in implementation depends on the internal public information effort to familiarize employees and other company personnel with the data center consolidation strategy that’s been created. It’s been observed that administering a data center consolidation strategy needs to be an “all-in” proposition, with all stakeholders informed about the strategy and what they need to do to support it. Effective change management can prevent strategy deviations, as could occur if staff members were to set up their own data centers independently of the new company policy.
What are the primary benefits of data center consolidation strategies?
Although there are numerous advantages to be gained by the organization that carefully implements a data center consolidation strategy, there’s one benefit that’s almost impossible for most companies to ignore: cost savings.
By reducing the number of data centers and other forms of outdated storage systems, an organization stands to substantially lower its operational costs. For example, a business will need fewer IT assets when it achieves consolidation, and this allows it to commence with the decommissioning of outdated equipment. By the same token, the company will need fewer data centers to warehouse data, and it won’t need to purchase new data centers. Each of these situations could generate a positive contribution to an organization’s bottom line by creating cost reductions like reduced maintenance costs.
The other benefits of having an effective data center consolidation strategy in place include improved performance, a streamlined management system for data storage, and enhanced data security with fewer security risks and outages due to the presence of centralized data assets.
One other benefit is a true “win-win,” both for the organization and the environment. When an organization adopts a data center consolidation strategy, that streamlining also increases the organization’s energy efficiency, which reduces the company’s energy consumption and lessens its carbon footprint.
Six steps for crafting a data center consolidation strategy
Typically, a data center consolidation process entails the following steps:
Step 1: Inventory data holdings
While there are different approaches recommended for handling a data center consolidation project, there’s little argument over what should be any organization’s first step—evaluating what data assets you have and confirming which data center facilities should house them. This needs to be performed with considerable precision and thoroughness given the profusion of data in most organizations.
Step 2: Define data centers
Before launching a data center migration process, an organization must first define the physical parameters of the data center(s). This begins with a study of the real estate involved and geographic aspects (such as physical size and available space). It should also include factors related to the data center’s electrical needs, including cabling, bandwidth, connectivity and the required power supply.
Step 3: Map the workload
The next step requires that the organization’s software and hardware configurations be mapped out so a fair judgment can be made about resources and usage. What’s working effectively? What isn’t? Having a well-defined discovery and dependency map assists this process, while techniques like virtualization help a company redistribute its workloads so that more workloads are handled by one machine.
Step 4: Assemble the team
Data center consolidation projects are not small endeavors. And given that they will affect each department within an organization, these projects must be led by personnel with surefooted project management. Likewise, a company should get some sense beforehand of what these various technical upgrades (like cloud migration) will end up costing before the full procurement budget process begins.
Step 5: Design the plan
At this point, an organization should have a comprehensive view of its data assets and a complete idea of how the organization’s new data storage system should work. So, it only remains for the IT architects to take all the assembled data about hardware and software systems and develop the final consolidation design. As soon as that design has been thoroughly vetted, the plan can be implemented.
Step 6: Test and confirm
Once an organization achieves a post-consolidation environment, it should take the time to fully test and confirm the viability of the revised data center infrastructure. That means checking all parts of the new system and ensuring the stability of the IT infrastructure.
Data center consolidation strategy best practices
Numerous best practices can be applied to data center consolidation:
- Monitor power usage: Data center consolidation is as much about power utilization as it is data. As a baseline to use in later comparisons, you need to understand how much energy is consumed by your existing configuration before data consolidation begins.
- Implement DCIM software: Data center consolidation is a mammoth job, requiring the use of data center infrastructure management (DCIM) software that monitors your IT equipment usage. DCIM software gives you the tools to run more efficiently.
- Assess assets and storage space: This is really a restating of the entire principle behind the development of a data center consolidation strategy. It all begins with thoroughly checking your data assets and comparing that with the space for allocation.
- Integrate automated processes: In order to reap the full benefits of data consolidation, it’s important to maximize your power efficiency. You can do that by automating power usage, which will help it reach peak efficiency by being controlled via automation.
- Eliminate deadweight hardware: A thorough inventory of your assets will likely reveal the presence of so-called “ghost servers” or “zombie servers.” These assets are contributing nothing and are in fact drains on electrical power and physical space.
Get started with your data center consolidation strategy
Successful data center consolidation provides numerous benefits to the organization that streamlines its data center operations. Having a consolidated data center lets you boost uptime while simultaneously reducing downtime. Beyond that, it’s an optimization process that aids in effective data center management. When used smartly in conjunction with existing business requirements, data center consolidation initiatives can prove to be the first step in effective infrastructure management.
Regardless of your business priorities, IBM offers the hardware and software solutions you need to safely store and protect your key resources—including storage management software that’s designed to help you consolidate data and do more with it.
Comments are closed.