Protecting your data center against power failure is crucial to providing maximum availability. Power loss or poor power quality is a major contributor to data center server down time. This task takes more than simply having UPS and a backup power source, such as a generator. Dual Power Paths: A Crucial Element in a High-Availability Data Center article tells that in in high availability environments, a common way to provide redundancy is to supply two independent power paths to each piece of computing equipment. All Dual Feeds Are Not Created Equal white paper explains the classification system for data center power feeds and provides the pros/cons of each configuration.
The use of dual power path architecture in combination with IT equipment with dual power supplies and power cords is an industry best-practice. Data centers designed and built utilizing Tier IV requirements are by definition “concurrently maintainable,” which means any system or component in the data center may be shut down for maintenance or may fail without affecting the delivery of services to the end user. In the case of a dual-powered data center, this typically is achieved by delivering at least two power circuits to each cabinet, one from the A power source and one from the B power source. The equipment accepts the two power feeds via independent, parallel power supplies that are sized such that the equipment will continue to operate with only one power path. To make this work the equipment to be powered need to have Redundant Power Supplies.
Redundant Power Supplies are one advanced feature available on high-end server machines. In essence, this is a power supply that actually includes two (or more) units within it, each of which is capable of powering the entire system by itself. If for some reason there is a failure in one of the units, the other one will seamlessly take over to prevent the loss of power to the PC. You can usually even replace the damaged unit without taking the machine down.
Design Tips for the Dual-Powered Data Center and Four A-B Design Failures to Avoid in the Dual-Powered Data Centerarticles gives some more tips how to design dual power paths. You need to be careful in designing dual-powered data center. Failure to properly design, size and implement dual power infrastructure at the cabinet may lead to breaker trip during restart (starting current of many computing devices and storage systems could easily exceed 200% of the running load for some time). You need to have enough spare capacity, but on the other hand failing to fully load power circuits to their rated capacity may not result in downtime but may drive power subscription costs higher than necessary. Proper power planning and budgeting involves loading every circuit to the proper rated capacity while respecting safety margins.
For dual power paths approach to be effective, you’ve got to meet two requirements:
- The protected equipment must support dual power feeds and operate with one feed faulted.
- The loading of breakers within each power path must always be less than 50% of trip rating during normal conditions, so the breaker doesn’t trip if the alternate path has to take on the full load.
Meeting these two requirements can be a challenge. Especially because some computing and networking equipment is only available with a single power cord. It’s good design principle to disallow the use of single power cord computer devices in a high-availability data center environment, but there are case where those can’t be avoided. For example some network products or legacy servers may only have single power supplies.
One good way to over-come single power input equipment problem is to use an Automatic Transfer Switch (ATS), which generates a single feed from two inputs. These single power supply devices can still be used with reliability by utilizing automatic transfer switches. Redundant Power Supply article has the following picture to show the use if automatic switch.
Powering single-cord equipment in dual path data center environments article tells about a new white paper from APC-Schneider Electric addresses the concern of powering single-corded equipment in dual path data center environments. However, equipment with a single power supply introduces a weakness into an otherwise highly available data center. Transfer switches are often used to enhance single-corded equipment availability by providing the benefits of redundant utility paths. You need to understand the use of power transfer switches well because there are several possible configurations how to use them, with their pros and cons.
Powering Single-corded Equipment in a Dual Path Environment white paper goes on to describe three fundamental approaches to powering single-corded equipment in a dual path data center environment. There are a number of options for integrating single-corded devices into a high availability dual path data center. Powering Single-corded Equipment in a Dual Path Environment paper explains the differences between the various options and provides a guide to selecting the appropriate approach. The conclusion is that Power availability to the single-corded equipment below 10 kVA is optimized by bringing utility redundancy directly to the rack. This can be done by using a rack mount static transfer switch or a rack mount ATS, and the optimal solution is a rack mount ATS.
A well designed adaptable rack enclosure power system would be able to support a single or dual path environment or a hybrid of both single and dual equipment. Automatic Transfer Switch (ATS) used in data center is typically rack mountable and occupy 1U or rack unit of space. They feature dual input cords and are able to switch from one power circuit to the other in a few micro seconds when power failure is detected on one of the input leads.
The idea to write about this topic to this blog came to me after reading Powering Single-corded Equipment in a Dual Path Environment white paper.