What is network convergence? Simple: the combination of multiple services over telecommunications infrastructure. Integrated pools of (virtualised) computing, storage and networking resources are increasingly shared across several applications, thanks to highly efficient, policy-driven processes. However, this introduces challenges for (enterprise) data centers...
Network convergence can also offer access to increasingly sophisticated system intelligence. It can deliver big (technical and business) efficiency increases. This includes full overview and control of physical infrastructure. Convergence with mobile computing and connectivity introduces new possibilities in, for example, IoT , Cloud, Digital Infrastructure management, commerce and finance. The next step will be new applications in areas such as Digital Enterprises, HD video and imaging, building automation, Smart City technology, Artificial Intelligence and Virtual Reality.
Data centers: more bandwidth, lower latency, same footprint
However, network convergence brings the need for maximum bandwidth and minimal latency. This has consequences for the data center. Increasingly demanding devices and cloud applications devices are likely to boost bandwidth requirements beyond 1 Gb/s. We will also see a vast increase in Machine-to-Machine and Device-to-Device communication, essential to developments such as smart cities (traffic management, video surveillance, mobility on demand), smart homes (energy management, automating, increased security), augmented and virtual reality, wearables, implanted technologies (health sector) and V2X communication. High performance networks supporting 2.5, 5, and 10Gb/s seem to be the most future-proof solution currently available.
Supporting network convergence
Rapid growth of data stored and managed in the data center requires the largest possible port density in the smallest possible space. Certain considerations need to be taken into account when creating an High Density solution. This kind of infrastructure requires more energy and produces more heat, for example, and racks can be become far heavier. Consolidating POP servers in a single rack unit leaves more space for switches and routers. Software Defined Network (SDN) architectures can be planned more sensibly. High density makes it possible to free up space for additional racks and switches and minimize the meet-me-room area.
Less space, more performance
Virtualisation requires less hardware to achieve the same levels of performance and manage the same workload. It supports the latest generation of energy-efficient, smaller data centre equipment. Thinner cabling for next-generation technologies means you don’t need to store as many cables in raised floors. This saves on materials and resources whilst improving air circulation – which saves energy. In addition, servers and switches need less transmission power thanks to shorter links.
Cables should have a very high fibre count, be easy to terminate and it should be possible to handle them in the same way as smaller cables. It’s important to note that bad cable management can result in inter-symbol interference, damage and failure, resulting in data transmission errors, performance issues and downtime. We recommend double-checking measurements and the quality of terminations, testing wherever necessary, labelling and colour coding, avoiding tight conduits and ensuring no cables or bundles resting upon others.
As the amount of end-user and device-to-device data grows exponentially, network quality, reliability and flexibility are key to current and future success. This requires an eye for detail in the execution and maintenance phases. Precisely specified requirements, in line with local needs and requirements, is key!