Vertiv announces ecosystem trends through 2018 data center trends report

Vertiv forecasted that a fourth-generation data center that intelligently and perfectly integrates mission-critical network edge functions beyond domain boundaries will emerge in 2018 and become the model for IT networks by 2020.

Vertiv recently announced five major 2018 trends related to the data center ecosystem, including the emergence of edge-dependent fourth-generation data centers, through its '2018 data center trends' report confirmed by the company's global expert groups.

Anand Sanghi, president overseeing Vertiv's Asia business, stated, "In the Asia region, the colocation facility sector has shown tremendous growth as enterprises have migrated core applications to the cloud over recent years. As this trend aligns with explosive growth in connected devices and demand for low latency and seamless customer experience, the need for architectural changes in mid-tier and edge data centers is being raised."

He added, "In 2017, Vertiv created over 50 customer use cases related to our converged/edge smart solutions in Australia, Singapore, Korea, and other regions, enabling enterprises to seamlessly integrate cloud to the edge and manage their infrastructure more agilely, efficiently, and visibly on a single screen."

1. Emergence of fourth-generation data centers: Both typical IT server rooms and 1,500 square-foot micro data centers are seeing increased organizational dependence on the edge. The fourth-generation data center comprehensively harmonizes edge and core integration, elevating this new architecture beyond a simple distributed network.

This is implemented in an innovative architecture that provides near-real-time capacity using optimized cooling solutions, high-density power supplies, lithium-ion batteries, and power distribution units (PDU) in economical scalable modules. By mobilizing advanced monitoring and management technologies, hundreds or thousands of distributed IT nodes can be operated simultaneously, enabling reduction in latency and initial costs, increased utilization, and elimination of complexity, allowing organizations to add network-connected IT capacity when and where needed.

2. Cloud providers moving to colocation: Rapid cloud adoption often causes cloud providers to fall behind capacity demand. In fact, some providers would be better off not attempting it. They prefer to focus on service delivery and other priorities rather than building new data centers, leaving capacity demand to be met by colocation providers.

Because colocation focuses on efficiency and scalability, it can quickly meet demand while lowering costs. With the proliferation of colocation facilities, cloud providers can choose local colocation partners that meet end-user requirements and operate them as edge facilities. Colocation responds by providing part of the data center for cloud services or providing entirely customized facilities.

3. Restructuring the mid-tier architecture of data centers: Everyone knows that the fastest-growing areas in the data center market are hyperscale facilities (primarily cloud or colocation providers) and network edge. As colocation and cloud resources increase, existing data center operators have gained the opportunity to reconsider and reconfigure their facilities and resources that remain important for local operations.

Organizations with multiple data centers are likely to continue consolidating internal IT resources, reducing scale by transitioning to cloud or colocation, and utilizing rapidly deployable configurations that can scale quickly when needed. While these new facilities will be smaller in scale, they will show improved efficiency, security, and availability, maintaining the mission-critical nature of the data these organizations seek to protect.

In regions where cloud and colocation adoption is slower, hybrid cloud architecture is expected to be the next step. This is an approach that combines private or public cloud with safer internal IT resources, reducing costs while managing risk.

4. High-density rack power: The data center community has predicted a surge in rack power density for the past decade, but in reality, the increase has been gradual. However, this situation is changing. While densities under 10kW per rack are typical, 15kW is not uncommon in hyperscale facilities, and some are gradually increasing to 25kW.

The reason this phenomenon is occurring now is primarily due to the adoption and widespread deployment of hyper-converged computing systems. Of course, colocation plays an important role in facilities, and higher rack density can increase revenue. Additionally, as energy efficiency performance in server and chip technology continues to advance, high-density adoption will likely follow soon.

However, the trend of major facilities moving toward high-density is expected to be rather gradual rather than rapid. If density becomes much higher, fundamental changes could come to the data center form factor, from power infrastructure to cooling methods for high-density environments. High-density will emerge, but it is likely to occur after 2018.

5. Global response to the edge: As more enterprises move computing to the network edge, there is a need for rigorous evaluation of the security and ownership of facilities containing these edge resources and the data within them. This includes complex questions about the physical and mechanical design, structure, and security of edge facilities, as well as data ownership. Governments and regulatory agencies worldwide will need to consider and respond to these issues.

Moving global data to cloud or colocation facilities and analyzing it again is too slow and cumbersome. Therefore, cases of placing data clusters and analytics capabilities at the edge located in various cities, states, or countries rather than at the central headquarters are increasing. While discussions continue about who owns the data and what is permitted to be done with the data, answers and measures for these discussions are expected to emerge in 2018.