What will happen to the future data center? (3) about future data center meetings

What will happen to the future data center? (3) about future data center meetings

Next I will talk about what will happen to the future data center? (1) and what will happen to the future data center? (3)"

In this article, we will analyze and examine the revolution and destructive force of the data center industry in the next decade with readers and friends, the new data center types and specific use cases will be predicted, and an effective method for future-oriented data centers to prevent downtime interruption and use innovation will be proposed.

Edge Data Center

In all the trends that shape the future development of data centers, the demand for edge computing is expected to be more important, so it deserves special attention from the industry. Edge computing can be described as the distribution of network edge in the vicinity of data generation and data usage. This may be an enterprise's factory workshop or operator's presence, mobile phone signal base station tower or smart building.

Although use cases such as content delivery network (CDN) and local processing and storage are expected to drive the demand for edge computing in the short term, IoT) it is expected to become one of the long-term driving factors for new edge capacity. There are many Iot use cases, but even in similar use cases, the data path and data center type are also different. We believe that many Iot deployments may eventually host data in a combination of public and non-public cloud facilities, and require distributed micro-data centers and very large centralized sites.

In some cases, major data, including the data required by other applications and staff, will be provided at the "near edge" of a large data center, the hosting of these large data centers and data centers in other cities are close to the data generation location. Cloud heavyweight enterprises are rapidly building ultra-large data centers and renting hosted sites through direct fiber links. This makes ultra-large scale cloud capacity closer to the edge, which can be effectively used as the "near edge" data center capacity. Cloud providers will also use the "Cloud"-distributed edge capacity for data caching or low-latency computing.

Once used or integrated, data is usually migrated or streamed to a large or very large remote data center for summary and analysis (including integration with other data and applications) and archive. These large facilities represent the "core layer ".

Future-oriented infrastructure

Some people think that the most efficient data center does not need to be built. Despite the benefits of new specifications, operators can use incentives to ensure that existing data center sites are available for as long as possible, with high productivity, high utilization and high efficiency. The 10-15 years of service life data center established today will also become a data center in the 1920s S and 1930s S. Running a hybrid IT environment provides greater flexibility in capacity management, but integrating or upgrading to a local IT infrastructure and implementing virtualization can improve the utilization and productivity of existing data center sites.

The IT hardware of the data center is usually updated within three to four years, but most of the infrastructure updates or renovations are more difficult and cost-effective. Over-configuring cooling and power overhead to meet future IT capacity requirements (balance with efficiency) is a standard practice. However, this can be described as future-oriented load changes, rather than new physical technologies that may lead to outdated design. However, emerging and established strategies and technologies, some of which are creating future data center types, can help to maximize efficiency and capacity management and reduce outdated risks. These strategies and technologies include:

• Modular Design-the use of PFM (including containers or micro data centers) to add new space can provide benefits to address the demand for short-term capacity, while also conducive to long-term innovation. Architects and Engineers can use presets to explore improved design methods for newer (sometimes radical) architectures, mechanical and electrical properties. Factory integration and modular installation of the next-generation PFM infrastructure will provide this new design for existing data center sites at a low cost and complexity.
• Flexible distribution and storage-use upgrades and intelligence (three-phase) certain technologies such as PDU, medium voltage distribution, and bus distribution provide scalability and adaptability to changes in workload requirements of existing data centers. Creating sufficient computing margin in a smart device (such as a smart PDU Controller) also helps to ensure its service life. Carriers will not test and use these technologies on existing data center sites, as they are likely to become increasingly standard in future data center construction. Emerging Energy Management, on-site power generation and storage technologies (including microgrid) can ultimately help future-oriented data center sites provide power availability and elasticity.
• Efficient and high-density heat dissipation-like power distribution, the flexibility of the Heat Dissipation design is crucial for effective capacity management and future-oriented protection. Currently, carriers dependent on mechanical cooling should study Non-cooler design, because free air cooling is likely to become the default choice for most places in the future (depending on the geographical location of the data center ). New capacity should also be created to support tightly coupled cooling, including backdoor heat transfer or direct liquid cooling (water needs to be allocated to the rack ).
• Management Software-DCIM and other tools can reduce risks, while achieving new efficiency, better capacity prediction and business agility. In addition, the use of DCIM and DMaaS to more closely monitor and manage the data of existing data center sites, promote analysis of insights, and significantly improve facility efficiency and capacity planning.

Conclusion and prospects

Today's data center operators are facing unprecedented changes. In the next 10 years, the data center situation may be in conflict with previous macro strength and technical trends:

• Today's distributed data center architecture may evolve into a more intelligent and highly interconnected network composed of new data center specifications. Compact, self-managed data center nodes and hubs (data or connectivity, or both) will be embedded and further popularized, supported by the Big Data Center Park.
• The new edge capacity will consist of significantly different data center types. They will include ultra-large scale cloud services and large host hosting facilities located near data generation points to support many applications; new micro-data centers on the edge; and smaller capacity clusters, these clusters may not be big enough or rigorous enough to be technically described as data centers.
• Large data centers can become their own power companies and act as energy centers. It can control the quantity, quality, and security of its power supply through private micro-grids, thereby improving energy elasticity.
• The convenience of connectivity, reliable power, and energy costs will continue to determine the site selection of new data center facilities. However, as data center operators respond to stricter regulations and public pressure, the importance of data governance and privacy will increase. The growth in edge demand driven by IoT and other applications/use cases will also drive new and special demands in major urban areas, as well as specification factors for network and business data centers.

It is foreseeable that, under the unprecedented revolution and new disruptive technology wave of the next decade, it is obvious that the data center will continue to promote innovation with all its new specifications and roles, it also promotes enterprises' adoption of emerging technologies and business transformation.