Agile connectivity and power are at the heart of today’s data center design.
Connectivity is reaching everywhere. Fixed and mobile Internet traffic volume and data-storage needs are growing at hard-to-comprehend rates. The Internet of Things (IoT) is promising more than 20 billion connected devices by the year 2020, according to Gartner. Cisco’s Visual Networking Index for 2016 projects that monthly global mobile data traffic will be 30.6 exabytes by 2020, increasing 8-fold between 2015 and 2020. The accelerating demand for fixed and mobile data dictates an increase in physical data center infrastructure, and this new infrastructure requires higher-speed connectivity and more electrical power. These needs are being met not only with new systems designs, but with next-generation data and power connectors that support forward-looking design requirements.
THE NEED FOR DATA AGILITY
Within the data center, data links are being pushed to ever-higher rates. Internal data lanes at 25 Gbps and now 50 Gbps are rapidly superseding lanes operating at 10 Gbps. There are single-lane approaches as well as use of multiple higher-speed links routed in parallel for yet-higher aggregate speeds – for example, a run of 4 x 25 Gbps lanes yields 100 Gbps overall performance.
Traditionally, designers need to route signals in many ways: from each line card or chassis to the topof-rack switch; to the end-of-row switch; from the front end-of-row switch to the core switch; and to an aggregation switch, to cite just a few of the many possible stages and transitions.
But now designers are routing signals in even more ways. Disaggregated architectures are driving innovative new ways of thinking about data flows in the new data center world.
Increased east/west traffic across racks and more server-to-server connectivity are driving leaf and spine architectures that need larger, high density switches and a greater number of overall internal and external ports. For many of these links across racks, within racks and even inside the box, copper wire cable solutions are proving attractive due to their moderate cost, ease of use, and performance.
Innovations in copper cabling are increasing capabilities and helping to deliver high-density signal and power with even greater efficiency. But there’s a difficult tradeoff to be made with data links. As the data rate goes up, the “reach” or achievable distance goes down.
A gain of speed may be outweighed by the need for repeaters, which extend the data-path distance, or by the requirement for a more tightly packed design. System architects must carefully consider their options to get the best combination of several factors, including size/weight, power consumption, cost and performance.
When upgrading existing systems, it is often easy to forget the cabling resources. However, to take full advantage of any hardware upgrades made, you need to consider upgrading your cable assemblies. Legacy cables for slower data rates likely won’t be able to deliver the performance required by the new speeds.
Agile connectivity and power are at the heart of today’s data center design.
Connectivity is reaching everywhere. Fixed and mobile Internet traffic volume and data-storage needs are growing at hard-to-comprehend rates. The Internet of Things (IoT) is promising more than 20 billion connected devices by the year 2020, according to Gartner. Cisco’s Visual Networking Index for 2016 projects that monthly global mobile data traffic will be 30.6 exabytes by 2020, increasing 8-fold between 2015 and 2020. The accelerating demand for fixed and mobile data dictates an increase in physical data center infrastructure, and this new infrastructure requires higher-speed connectivity and more electrical power. These needs are being met not only with new systems designs, but with next-generation data and power connectors that support forward-looking design requirements.
THE NEED FOR DATA AGILITY
Within the data center, data links are being pushed to ever-higher rates. Internal data lanes at 25 Gbps and now 50 Gbps are rapidly superseding lanes operating at 10 Gbps. There are single-lane approaches as well as use of multiple higher-speed links routed in parallel for yet-higher aggregate speeds – for example, a run of 4 x 25 Gbps lanes yields 100 Gbps overall performance.
Traditionally, designers need to route signals in many ways: from each line card or chassis to the topof-rack switch; to the end-of-row switch; from the front end-of-row switch to the core switch; and to an aggregation switch, to cite just a few of the many possible stages and transitions.
But now designers are routing signals in even more ways. Disaggregated architectures are driving innovative new ways of thinking about data flows in the new data center world.
Increased east/west traffic across racks and more server-to-server connectivity are driving leaf and spine architectures that need larger, high density switches and a greater number of overall internal and external ports. For many of these links across racks, within racks and even inside the box, copper wire cable solutions are proving attractive due to their moderate cost, ease of use, and performance.
Innovations in copper cabling are increasing capabilities and helping to deliver high-density signal and power with even greater efficiency. But there’s a difficult tradeoff to be made with data links. As the data rate goes up, the “reach” or achievable distance goes down.
A gain of speed may be outweighed by the need for repeaters, which extend the data-path distance, or by the requirement for a more tightly packed design. System architects must carefully consider their options to get the best combination of several factors, including size/weight, power consumption, cost and performance.
When upgrading existing systems, it is often easy to forget the cabling resources. However, to take full advantage of any hardware upgrades made, you need to consider upgrading your cable assemblies. Legacy cables for slower data rates likely won’t be able to deliver the performance required by the new speeds.
