The Sweet 16 Factors of Data Center and Cloud
You may have heard a little something about Sweet 16 this week.
Local NCAA Basketball tourney qualifiers Villanova, Penn and Lehigh lost last week, as did the Villanova women’s team. Congrats to the Quakers and Explorers for winning the Ivy and Patriot Leagues respectively, as well as to both Villanova teams for scoring at-large bids.
With our local teams now out of the tournament we want to suggest a different Sweet 16.
Let’s review a collection of 16 important factors for you to assess when making decisions on your IT infrastructure strategy and your mix of data center, cloud, networks, and associated services.
These are the Sweet 16 factors of data center infrastructure and cloud services:
Resilience: To what level of resilience is your data center designed for and operating at? Are there any single points of failure? Are your backup systems up to date and tested regularly?
Uptime History: Have the design and operations of the data center been executed as intended, leading to an impressive level of uptime? The “five nines” standard of 99.999% or higher is a good measuring stick. Being down is bad for business, and for your career. The very sophisticated, complex infrastructure of public cloud providers is a risk factor that has led to them having spotty uptime records.
Cost Control: IT infrastructure is a major cost item for most organizations; how do you keep it from being a budget buster? Cloud makes a lot of promises and it’s scalability can be a great tool for companies who are prepared to manage it, but the utility model of public cloud makes it very difficult to control costs even for companies with top talent and a good plan.
Interconnection and Network Options: Do you own and operate a data center with only a carrier or two? As the menu of carrier product options expands, that may prove to be limiting. A carrier-neutral, fiber rich colocation provider with numerous options gives you more flexibility and choice, especially when a carrier has significant network issues that need to be overcome.
Physical connectivity resilience: Having multiple carriers enhances resiliency, but ideally those carriers enter a data center through multiple physical locations to provide resiliency. Coupled with multiple carriers having multiple pathways and you’ve avoided having any single points of failure in your network.
Cloud options: Even those enterprises focused on security, performance, and reliability will typically implement cloud applications for back office and non-mission critical functions. So, most companies pursue a hybrid approach utilizing both the cloud and traditional data center infrastructure. Standard internet connections can create a bottleneck for these hybrid strategies. A data center with cloud onramps enables a true hybrid approach.
Compliance Support: The evolution of requirements to achieve regulatory certifications will always be with us. Access records, record-keeping, redundancy, are but a few of the factors that go into achieving and maintaining the compliance standards your customers and partners require.
Security: Fully staffed 24x7x365 security is a must. This goes beyond just having a guard at the desk; biometric access control, escort requirements, careful monitoring of interior and exterior security cameras are all requirements of a highly functioning security plan.
Service & Staffing: If you need help with something, it is best to be able to get someone knowledgeable, helpful, and experienced on the phone. An on-site, experienced company representative readily available is best. An overseas call center or trouble ticket system with slow response time, not so much.
Location: If you’re in the public cloud you never truly know where your data is, other than in one of the provider’s regions. If you’re using traditional data center infrastructure it needs to be convenient, safe, and outside of flood zones and other high risks of natural disaster.
Scalability: The cloud is great for scalability but get ready to pay for the privilege. For predictable workloads, the economics are nearly always better to run them on traditional data center infrastructure. Remember that scale goes two ways, both up and down. Corporate data centers built for requirements from 15-20 years ago likely have stranded capacity. If your needs for traditional data center infrastructure grow or diminish over time colocation allows for adjustments with each contract renewal.
Flexibility: In an era of changing needs, the opportunity to pursue flexible solutions is paramount. Your business needs may lead to a rapid shift in requirements for reasons of technological change. Or the need for flexibility may have nothing to do with technological advancement, but can be due to corporate M&A activity, workforce consolidation, pivots in financial strategy, mandates from the C-level and the board, budget adjustments, and many other factors which can take you from aggressive usage of the public cloud to more reliance upon traditional data center infrastructure.
Partner/Customer Preference: You may have partners or key customers that prefer to interact with you on a specific infrastructure for a joint venture, for a project or product offering requiring low latency, or something else you’re doing collaboratively. This is not as rare as you may think, especially among more technologically sophisticated organizations.
Managed/Technical Services: You may have specific needs for features or tools that are best delivered by a managed services or technical services provider. Some cloud and colocation facilities bundle one-size-fits-all white-labeled managed services while many colocation providers have a variety of providers on site and fast fiber interconnections, creating a low-latency ecosystem for potential partnerships and vendor/customer relationships.
Real Estate Flexibility: Pennsylvania data centers built for an organization’s requirements from 15-20 years ago may have stranded capacity that leads to a white elephant data center, a real estate liability. Increasing hybrid and work-from-home employment has also resulted in enterprises looking to downsize their physical footprint. Putting your IT infrastructure into a colocation facility provides flexibility that doesn’t exist with data center ownership.
AI Readiness: Artificial Intelligence workloads are extremely expensive in the cloud and require power and cooling resources beyond the capabilities of most existing corporate data centers. Also, many enterprises are reluctant to upload their confidential and mission critical business data into a cloud-based AI. Quality colocation data centers can meet the power density requirements for AI and high performance computing.
Your requirements are complicated and change frequently. As a result, your strategy is likely to evolve sooner rather than later as well. If you’d like to have a conversation about your changing infrastructure needs and how Direct LTx’s hybrid colocation solutions can help you meet them, let us know.
Email us at strategy@DirectLTx.com for a conversation, a data center tour, or both, and enjoy the rest of the tournament.