Server Labs: Data center power availability provisioning

Data center power availability provisioning

Data center power availability provisioning

Through my previous post you would all have understand the importance of having power redundancy to provide maximum uptime. However we do have to consider that this engineering is designed as per our requirement as conservative over-provisioning of power capacity for availability reasons results in significant capital investment for large enterprises because this capacity is designed for failure conditions that do not happen often. On the other hand, under-provisioning this capacity runs the risk of affecting the performance of the data centre when failures do happen, through either service unavailability or degraded service performance. Hence, there are interactions and trade-offs between power capacity utilization, power redundancy, and data centre performance that are often overlooked.

Let us discuss the four ways for power provisioning and you can decide the best solution required for your data center.

Power provision - Single Path
Power provision - Single path with resilience
Power provision - Concurrently maintainable
Power provision - Fault tolerant

Power provision– Single Path

Basic data centre with non-redundant capacity components and a single non-redundant distribution path serving the computer equipment. This configuration could be described as a Tier I or rated 1.

This distribution path incorporates:

A single main electrical supply from the utility
A distribution network to split the critical load from the non-critical load. The common belief here is the critical load is the IT load, and everything else is non-critical.
A UPS of some description to ensure the supply to the critical load is maintained.
A standby system (usually a generator) to replace the utility supply in the case of a power failure
Considering that this model is single points of failure that would cause the data centre to become unavailable in case of any downtime for electric power.
Planned maintenance work will require most or all of the facility infrastructure systems to be shut down, impacting networking/computer operations.

Power provision – Single path with resilience

A data centre with some redundant capacity components and a single, non-redundant path serving the computer equipment. This would equate to a tier II or rated 2.

The distribution path incorporates:

A main electrical supply from the utility.
A distribution network to split the critical load from the non-critical load. The common belief here is the critical load is the IT load, and everything else is non-critical.
Duplicated UPS of some description to ensure the supply to the critical load is maintained.
Duplicated standby systems (usually a generator) to replace the utility supply in the case of a power failure.
Redundant capacity components that can be removed from service on a planned basis without causing shutdown to active IT environments.
The additional capital cost of the duplicated components is self-evident, but there is also an ongoing running cost penalty as the systems are running at low utilizations by design.

Power provision - Concurrently maintainable

A data centre with redundant capacity components and multiple independent distribution paths serving the computer equipment. This would equate to a tier III or rated 3.

The distribution paths incorporate:

Diverse electrical supplies from the utility.
Duplicated distribution networks to split the critical load from the non-critical load. The common belief here is the critical load is the IT load, and everything else is non-critical.
Redundant UPS of some description to ensure the supply to the critical load is maintained.
Redundant standby systems (usually a generator) to replace the utility supply in the case of a power failure.
Any of the capacity components and elements in the distribution paths can be removed from service on a planned basis without impacting active IT environments.
Only one distribution path is required to serve the computer equipment at any time.
The additional capital cost of the duplicated systems is self-evident, but there is also a very significant ongoing running cost penalty as the systems are running at very low utilization, by design.

Power provision - Fault tolerant

A data centre with multiple, independent, physically isolated systems providing redundant capacity components and multiple independent, diverse and active distribution paths simultaneously serving the computer equipment. This would equate to a tier IV or rated 4.

The distribution paths incorporate:

Diverse electrical supplies from the utility.
Duplicated distribution networks to split the critical load from the non-critical load. The common belief here is that the critical load is the IT load, and everything else is non-critical.
Duplicated UPS of some description in each network to ensure the supply to the critical load is maintained.
Duplicated standby systems (usually a generator) in each network to replace the utility supply in the case of a power failure.
The redundant capacity components and diverse distribution paths are configured such that “N”.
capacity is providing power and cooling to the computer equipment after any infrastructure failure.
Each and every capacity component and element in the distribution paths can be removed from service on a planned basis without impacting active IT environments.
A single failure of any capacity system, component or distribution element will not impact the operation of the computer equipment.
All components must be compartmentalized (tier IV) and there is a requirement for continuous cooling (tier IV).
The additional capital cost of duplicated systems is self-evident, but there is also a very significant ongoing running cost penalty as the systems are running at very low utilization, by design.