Srikanth
They say that content is king. It helps to drive growth of network traffic and the pace is absolutely amazing. And, at the focal point of the trend is data center. A recent study has revealed that there will be growth till 15 zettabytes in network traffic by the year 2020, with more than 27 percent CAGR. With the growth of apps and technologies online, there will be an intensification of traffic volumes, in 2/3. Alongside the growth of data, the managers of the data centers will have to face the challenges regarding interoperability, rise in costs, complicated multi-fiber infrastructure, etc.
The product of the operators is their network, and extensive tests are conducted, almost each time they have some new network to introduce. The mindset of the ICPs is different. Content is their product, and they consider the network a channel for reaching the users. Limited tests are conducted by them prior to new services getting launched, as it is time-consuming for them, and this is a hindrance in the growth. Understanding this mindset of ICP and the impact of it on the infrastructure of network along with the necessities is very needed for the continuation of the rapid expansion that the data centers are going through.
Increasing automation
Telecom operator teams have several engineers at their disposal, who are involved in the management of the networks, and have proficiency in hardware as well. They take the hardware lifespan into consideration, alongside the technology for a span of 10 or more years. The operational teams of ICPs are smaller, while their work involves routing as well as software. They make APIs, as well as automation based on software, for the maximization of workloads as well as networks. The growth rate seen by the infrastructure of ICPs is phenomenal, which requires the technology to be ripped out and be replaced at a gap of 3-5 years, and this helps the hardware to be viewed for a shorter lifespan. There has been great growth at a rapid rate seen by ICPs, and agility is necessary now, in an extremely fast world. However, for the infrastructure of data centers, these can lead to downtime and interoperability problems. As if such challenges did not suffice, the data centers need to prepare themselves to deal with greater network speeds, too.
Maximization of speed, reduction of power
The speeds at DCIs as well as intra-connects have reached 100G already, and they will soon, the norm will be 400G. However, with the increase in speeds, managers of infrastructure will be required to maintain the same momentum, as they stay within the constraints of power. A major issue that data centers have to face is the reduction of the consumption of power in their infrastructure, as connectivity at high speed is delivered, and data demand at a growing rate is fed. It has been found that 400 TWH of electricity was consumed by data centers across the world, and this is much more than the consumption of the UK. It could become thrice in the next decade. As the data centers get higher pressure for the reduction of energy consumption, colder climates have started to see like a good option for some ICPs. Kolos, a joint venture by the US and Norway, is at work on the biggest data centre of the world at Arctic Circle, hoping for a 60% reduction of costs.
Testing the Limits
With the continued expansion of the ICPs, there will be an increased number of data centers for the accommodation of the rise in content level, and need seamless DCI, for the delivery of services to the users at a much higher speed. Considering the rapid pace of growth of these businesses, there has been hardly any time for the ICPs to arrange rigorous procedures that need to be tested, for the seamless DCI, which is a big challenge that many managers of data centers have to face, especially as they struggle with the rise in costs of infrastructure of cabling, alongside the several protocols of interoperating. Such challenges may look enormous. However, some steps taken by the managers help to keep things steady, and these steps are in the areas of test as well as measurement within data centers.
