How to assess the carbon footprint of your storage arrays
Monday, 28 November, 2011
With the government finally putting a price on carbon, IT managers around Australia will now have to pay close attention to the efficiency of their storage arrays. Clive Gold, Vice-Chair of SNIA ANZ, explains the rationale behind the association’s new Emerald program, and how you can use it to assess the impact of the carbon price on your own equipment.
‘Green’ is back on the agenda in Australia, now that we are looking at a cost for carbon at $23 per tonne for the next three years. The question is: how will this tax translate to the cost of running IT?
In the last few years there has been a great deal of work done to classify the power consumption of servers. Today you can get specifications for most servers and use these to understand their power draw when idle, and when processing. However, there has been no industry-validated way to measure and compare the energy efficiency of storage arrays. Now, as data volumes grow, storage continues to contribute more to overall data centre power consumption and customers need tools for consistent measurement.
In the past this has not been an easy thing to do. The goal is to have a measure of how many GB of storage you can have per watt of energy. The underlying issue is the service level and the loading on that storage. There are many variables that affect this, such as:
- Storage media used
- Structure - RAID level
- Access profile - read/write
- Efficiency features - deduplication, compression, etc
SNIA Emerald aims to provide such a measure. It is the result of a number of years of work by SNIA with both the industry and the authorities - such as the EPA (Environmental Protection Agency) in the USA - to define a meaningful measure, and describe a reliable and consistent testing methodology. The original concept was similar to the star ratings which you get on household appliances. With this simple measure, customers can factor energy consumption into their purchasing decisions, allowing them to weigh up a device’s features and functionality, while also being able to estimate how much its total running costs will be.
Due to the complexity of the problem, the industry has not been able to formulate a simple measure like the star ratings found in consumer technology. However, SNIA has managed to distil the issue down to a few pages, to help customers understand the energy consumption of their arrays and estimate how much power they will consume.
The data is freely available at www.SNIAemerald.com, and although there are only a limited number of completed tests available today, we hope that over the coming months, all vendors will help to complete this database. Storage system manufacturers can download the SNIA Emerald Power Efficiency Measurement Specification from the SNIA Emerald website, as well as a User Guide that provides step-by-step guidance on how to set up a test and measurement environment for a storage system under test, and then submit measured test results to the SNIA Emerald Program. Once submitted test results are approved for public posting, manufacturers will obtain a SNIA Emerald Program logo to highlight their program participation. The industry at large can then view the posted test results of various storage systems and review products that underwent the SNIA Emerald testing requirements.
Although in general Australian IT managers are not responsible for the energy consumption in the data centre, with the new tax this is probably going to change. After all, we are being told that the reason for the tax is to change behaviour. So next time you are evaluating storage arrays, have a look at the Emerald site and if your vendor’s storage array is not there, ask your vendor why not!
Is the Australian tech skills gap a myth?
As Australia navigates this shift towards a skills-based economy, addressing the learning gap...
How 'pre-mortem' analysis can support successful IT deployments
As IT projects become more complex, the adoption of pre-mortem analysis should be a standard...
The key to navigating the data privacy dilemma
Feeding personal and sensitive consumer data into AI models presents a privacy challenge.