Last week we launched a special report series on the new cloud and data center balance. This week, we’ll look at design considerations to help you create a more balanced cloud architecture.
Working with digital infrastructure solutions, specifically those designed to support cloud and data center operations, can be challenging. Growth around distributed computing, more requirements around application resiliency, and ensuring optimal uptime have all placed new stressors on data center and technology pros. This is all a part of the greater digital transformation effort. A recent report indicated that Digital Transformation spending would grow to more than 53% of all information and communications technology (ICT) investment by 2023, up from 36% today. Funds continue to pour into Digital Transformation efforts as business leaders see the category as a long-term investment. As we look ahead, further spending will approach the $2 trillion mark by 2022. But where do data centers and our digital infrastructure fit in with all of this? And, where are some challenges they face with all of these modernization efforts?
Emerging challenges around cloud; efficiency, resiliency, safety, and latency
We often get so excited about our digital infrastructure efforts that we don’t always see the potholes on the roads. Emerging challenges are complicating our deployment strategies. Although frustrating, professionals working in the technology space and those trying to balance cloud and data center solutions often encounter these challenges.
Keeping things modern and efficient
When it comes to the innovation part of the cloud and data center balance, complacency and a fear of testing new technology will impact your ability to be a part of the digital economy. A challenge often seen within data centers is the approach to modernization as only a partial effort. Even if everything does not need to be updated, you still need to review each component in your data center. For example, are you leveraging the latest high-speed copper and fiber cabling connecting switches to servers within edge data center cabinets? Or, are you using new solutions around power distribution, environmental, and connectivity hardware to improve availability, scalability, power, cooling efficiency, and product quality? Remember, these considerations not only help you create the right type of infrastructure balance they also minimize network downtime and increases data center productivity. Keeping this modern and efficient also involves a look at processes as well as technology.
Overcoming Latency and Connectivity Issues
Latency, connectivity challenges, and flat-out outages are incredibly detrimental to business operations. This is why network technologies must continually adapt. To the network infrastructure professional, that means an ongoing effort to support exponentially increasing application speeds and bandwidth requirements with the capacity and flexibility to efficiently deliver critical network services to the enterprise. But even as network technologies advance at a breakneck pace, one thing remains constant: The LAN is the backbone of the enterprise, and the cabling infrastructure is the backbone of the LAN. There is a further focus on the LAN for the modern data center as systems become more converged and efficient.
Further, LANs are being tasked with handling more traffic from various solutions, including hyperconverged architecture, high-performance computing, and more. Remember, not every cable is made the same, and not every networking solution can handle high-performance, high-density, and reliable communications. Working with network and connectivity in the data center space requires good partners and the right solutions.
Here’s an example. Emerging Category 6A UTP copper cabling systems now utilize innovative technologies directly in the cable and patch cords along with advanced connector compensation techniques to achieve channel bandwidth performance above industry standard requirements. These unique cabling solutions are constructed of discontinuous metallic elements that provide a high degree of alien crosstalk suppression. Further, these cables are smaller with helps with deployment and bundle size. These Category 6A UTP cabling systems also excel in performance without the need for bonding and grounding. They also eliminate the concern for shielding current flow arising from ground potential differences.
Managing an Overwhelming Amount of Complexity and Fragmentation
The more connectivity we work with and the more users connecting into our infrastructure, the more complicated it can become. Many deploy the latest and greatest to find themselves looking at more screens, checking more sensors, and examining even more data even with modernization efforts.
This is why you need to look for systems that provide easy-to-use navigation that renders powerful hierarchical and subcomponent diagrams of data centers, IT inventory, and networks. You need to have visibility, control, and governance to improve capacity planning and energy consumption to reduce complexity. Furthermore, a reduction in complex systems helps administrators isolate troubleshooting and minimize downtime for improved service levels.
As a new challenge facing many cloud providers, leaders in the cloud space suddenly face a boom in demand but have inefficient methods to deploy new infrastructure. Some of the largest cloud providers are actively looking at ways they can improve their speed-to-deployment and time-to-market. They’re doing this to support new edge requirements, new applications that require latency controls, and more connected users leveraging cloud solutions. The challenge has been the legacy focus on large, centralized ecosystems. Although these cloud solutions are still essential, they’re missing the mark in going after a market that’s already booming. This is a reason why cloud providers are pivoting from the early focus on building these massive centralized data centers to now partnering with colocation companies for regional coverage to improve latency, sustainability, and time-to-market approaches.
Incorporating Safety into a Physical Design
Cloud and data center solutions must be safe and secure. Those in the cloud and data center space working towards the right balance actively architect around safety from the beginning. By utilizing a Prevention through Design (PtD) approach, leading technology partners are developing a groundbreaking new electrical safety technology to help build a safer industrial infrastructure. From automating the absence of voltage verification process to “no tool” circuit breaker lockout devices, leading data center partners have created safety and security solutions that provide increased risk reduction from the front gate to the facility’s back without sacrificing productivity.
Here’s another critical challenge often faced when designing cloud and data center ecosystems. When it comes to a power distribution system, the potential damage from a short circuit fault can be catastrophic — resulting in an arc flash event at a piece of equipment or power cables whipping about violently.
These types of events pose a serious safety hazard to anyone nearby when it occurs, either during the initial power-up or afterward. This is why, to protect electrical infrastructure when using power cables in a tray, installing an IEC 61914:2015-compliant cable cleat is the best option when facing a peak short circuit fault.
This is quite a bit to take in. However, it’s essential to now dive into some design considerations as it elates to digital infrastructure. After reviewing emerging trends and some of the associated challenges, let’s examine real-world solutions that help create a more solid cloud and data center design balance.
How digital infrastructure and data center designs have changed
Since the dot-com bubble, data centers have only continued to grow in size and importance. They’ve also changed in how they’re being designed and delivered for digital solutions, new types of workloads, and of course, the cloud.
Building a Data Center for the Future
The rate of changing business demands and the explosive growth of the amounts of data being transmitted today require that data centers and cloud ecosystems be “purpose-built,” based on business direction and long-term growth projections. This calls for an architecture that enables fast and efficient scalability to meet capacity needs or migrate to higher data speeds. The pressure to sustain this pace is putting increasing strain on both IT and Facilities. Emerging Converged Infrastructure Solutions deliver a complete family of data center hardware, software, and services. These solutions are modular, pre-tested, and proven to work together, allowing IT and Facilities stakeholders to stay ahead of demands, minimize required resources and expenses and reduce the total cost of ownership. This approach helps to:
- Build or refine data centers to address changing IT and business demands
- Simplify and accelerate data center design and deployment of physical to logical convergence
- Reduce total cost of ownership by increasing operational efficiencies
- Minimize integration issues
- Rely on a primary vendor backed by a large ecosystem of world-class partners
By deploying a comprehensive yet complementary suite of hardware and software, supported by an expert services team who understands them intimately, installation can be executed quickly, efficiently, and with quality assurance. This translates to faster deployment speeds and an improved time to market for business operations. For cloud providers aiming at new markets, this is a great way to alleviate their growth challenges and deliver intelligent infrastructure to support their centralized cloud while improving networking and latency control capabilities. Let’s pause here and focus on a design consideration.
By deploying a comprehensive yet complementary suite of hardware and software, supported by an expert services team who understands them intimately, installation can be executed quickly, efficiently, and with quality assurance.
Cloud and data center experts need efficient, fail- proof cabling and the underlying infrastructure to operate at maximum uptime. As your business needs grow, you also need a way to seamlessly migrate to higher data rate speeds (40G/100G or beyond). Professionals in the data center architecture space provide the products and services needed to support uptime requirements for the most complex environments, including those that employ software-defined networking, application-centric infrastructures, and two-tier network providers. Look for solutions that include:
- Modular cabinets for serviceability and flexibility for moves adds and changes
- Seamless migration path for higher data rates and next-generation network architectures (fiber and copper connectivity)
- Passive airflow remediation product suite (thermal management, cooling control, and containment)
- Pre-configured cabinets, including plug-and-play solutions and POD containment designs
- Security, safety, and remote operational capabilities
Although we’ll discuss your capability to address a digital economy, one of the significant differences when looking at legacy data center solutions versus those deployed today is the ability to increase your time to market and improve deployment speeds. Consider this, pre-configured cabinet solutions can reduce time-to-production by up to 80 percent. This translates to faster builds, more efficient deployments, and a better competitive stance in a digital market.
Infrastructure Visibility and Management
A significant change in how we control today’s most advanced cloud and data center environments revolves around visibility, monitoring, and management. New DCIM solutions, used by some of the most prominent vendors in the industry, are designed to add intelligence to the data center, using dynamic solutions to derive actionable information about equipment locations, connectivity points, power usage, and power capacity. Armed with this information, organizations can identify areas for ongoing optimization of data center operations.
DCIM solutions serve as a “bookend,” working cohesively together at each end of the physical infrastructure. At one end, DCIM solutions bridge the IT stack by delivering visibility into asset and connectivity management to help streamline capacity management efforts and accelerate work order management tasks. At the other end, DCIM bridges the facilities stack by monitoring power usage and cooling efficiencies to help drive operational effectiveness and data center uptime.
For advanced data center design, look for DCIM software solutions that include a DCIM software suite, available separately or bundled, that includes individual modules for:
- Power management
- Thermal management and cooling optimization
- Asset and connectivity management
Similarly, look for intelligent hardware offerings that include:
- Monitored rack PDUs that provide management down to the outlet level
- Environmental monitoring, including wireless nodes that use a mesh network that is self-configurable, self-healing and non-invasive
- Intelligent patch panels that offer port-level connectivity traceability
- Pre-configured cabinets that offer DCIM plug-and- play benefits for day-one interoperability and performance
- Security and remote operations
Remember, in a digital economy, an outage or any extended downtime is very costly. And that cost only continues to rise. According to Gartner, the average cost of IT downtime is $5,600 per minute but can be as much as $540,000 per hour. This is why it’s essential to have direct visibility into all aspects of your data center operations. It becomes even more critical as your cloud and data center system become further distributed.
A New Kind of Supply Chain and Partnership
Having locally based professionals is a significant advantage that leading data center partners leverage for understanding various cultures, codes, and standards, all intending to provide a solution that suits your need in any given geography.
This is why it’s critical to ask questions and challenge your partners on their supply chain management. As an essential point regarding supply chains, look for partners with an extensive worldwide network of local distributors and regional warehouses. A significant consideration for those looking to better balance cloud and data center operations is a partner’s ability to grow beyond traditional centralized ecosystems. A good partner that will help you improve your time-to-market will also help you expand your regional coverage.
These types of partners allow you to deliver the right solutions when and where they are needed. Having locally based professionals is a significant advantage that leading data center partners leverage for understanding various cultures, codes, and standards, all intending to provide a solution that suits your need in any given geography. In short, select suppliers that operate globally while excelling at local execution; high touch account management, local contractors and installers, purposeful deliveries that may require regional manufacturing, and in-market inventory.
It’ll be essential to have alliance partnerships with industry-leading technology companies that enable you to design state-of-the-art solutions.
As customer needs and technologies evolve, your partner needs to be prepared for tomorrow. It’ll be essential to have alliance partnerships with industry- leading technology companies that enable you to design state-of-the-art solutions.
Controlling Latency, Enabling Performance
There have been significant changes in delivering data, applications, and workloads between cloud and data center ecosystems. As data center convergence becomes more commonplace, so do the demands placed on the physical infrastructure.
Emerging High-Speed Data Transport (HSDT) systems can help, whether you’re trying to:
- increase your network’s throughput
- reduce latency
- enable improved service levels
From a broad perspective on latency control and enabling performance, look for an offering that includes a comprehensive set of end-to-end HSDT solutions, including:
- optical fiber
- direct attach copper cable assemblies
- twisted-pair copper cabling
This gives data center architects and managers maximum flexibility in designing, layout, and implementing their data centers.
Remember, HSDT enables advanced network designs such as fabric-based architectures and FCoE and higher bandwidth deployments such as 10/40/100 Gig Ethernet LANs and 8/16/32 Gig Fibre Channel SANs to be deployed independently or in various combinations. HSDT also works with pre-configured infrastructures, which lowers speed-to-deployment, enhances thermal performance, decreases energy usage, and reduces the total cost of ownership.
Download the full report, “Performance, Efficiency, and Sustainability: The New Cloud and Data Center Balance” courtesy of Panduit to learn more about how the new cloud and data center balance. In our next article, we’ll look at solutions for cloud business leaders who are constantly looking to improve time to market, specifically, how to find a partner offering solutions that are both highly available and easily deployable. Catch up on last week’s article here.