As anyone involved in data center management knows, growth in network traffic demand and service is only increasing exponentially. In order to keep up with future demand, this means data center managers must formulate a proactive strategy for data center evolution.
Due to frequent and rapid changes, data center managers often find themselves reacting to events and crises rather than implementing a proactive strategy. Nevertheless, formulating that strategy is essential. With the number of devices connected to IP networks predicted to be three times higher than the global population by 2019, data center managers must always look to the future to anticipate necessary evolutions.
So, what should data center managers prioritize as they formulate a proactive strategy for the evolution of the data center?
There are three key areas that, we believe, should be part of a successful strategy: Migration to higher speeds, infrastructure management, and cloud performance. We’ve already explored strategic considerations for migration to higher speeds and infrastructure management. This post outlines the considerations around cloud performance.
As the trend towards virtualization and cloud deployment continues to gain momentum, any data center strategy will need to adapt to these developments. Indeed, the cloud platform and infrastructure market generates roughly $21 billion in revenue and is expected to grow 20 percent a year. Estimates also predict that cloud applications will account for 90 percent of worldwide mobile data traffic by 2019.
As a result, physical layer infrastructure will heavily determine the ability to effectively migrate on-premise applications to either private or public cloud so it’s crucial to plan carefully. It’s not simply a matter of ripping out old hardware and replacing it with new hardware; rather, it is an application landscape redesign.
To maximize the advantages of cloud-based computing, the platform needs to deliver low latency throughout; this highlights the importance of the flattened, two-tier architecture of the spine and leaf network. This approach features a high density of fiber links between the spine and leaf layers. Having high-or-ultra-high-density fiber panels is also important; you should opt for a connector design that ensures the highest possible utilization of every fiber strand.
Data center managers should also consider that the more fiber dense an environment is, the greater the need for low-profile cabling and connector modules, like the micro quad small form-factor pluggable (microQSFP). MicroQSFP ports offer the industry a familiar module – they are the same width as the existing single-channel small form-factor pluggable connectors – yet offer four times the data capacity. The infrastructure must also scale easily while keeping the increased fiber density manageable. The progression of higher density multi-source agreements (MSAs), as per the 2016 Ethernet Roadmap, emphasises the importance of a well-designed and flexible routing solution.
Ultimately, following a strong data center strategy will result in a data center environment which is simple to manage, efficient to operate, and easy to scale. This presents several benefits, including lower operational costs, faster turn-up for new and upgrading services, accelerated meantime to repair, and reduced downtime; all of which result in increased revenue for the organization.
When creating a data center strategy, it’s crucial to understand how migration to the cloud will impact data center design and performance – and make sure that you have the right physical layer infrastructure in place.
You can also learn more about considerations for the modern data center here.