On Amazon.com web services blog, there is a graph of bandwidth consumed by AWS vs. Amazon's web sites.

Lots of Bits

In January of 2008 we announced that the Amazon Web Services now consume more bandwidth than do the entire global network of Amazon.com retail sites.

Amazon.com CEO Jeff Bezos has been showing a chart of the relative bandwidth usage and I just received permission to post it here:

 

Amazon's Web Services growth has created a new force in online services. And Amazon's secret to its success may be its customer service focus.  Amazon is like a Nordstrom style retailer in that the customer is right.  How else can you explain amazon's rapid growth vs. the competition.

WSJ has an article highlighting the difficulties in new power plant construction.

Construction costs for power plants have more than doubled since 2000, according to new index data to be released Tuesday, and inflationary pressures will continue to put the squeeze on electricity prices.

The findings are bad news for consumers and utilities alike, and help explain why power-plant development has become something of a quagmire in the U.S. -- with no type of plant emerging as a reasonably priced option that can meet rising demand for electricity.

The analysis comes in the form of a price index from Cambridge Energy Research Associates Inc., a research and consulting firm in Massachusetts that is a unit of IHS Co. Similar to the consumer-price index, it calculates the cost of building new power plants based on the cost of materials and other factors.

"Costs for labor, materials, equipment and design and engineering -- all are up," said Candida Scott, senior director of cost and technology for CERA. As a result, the cost of building new plants is up 19% from a year ago and up 69% from 2005.

The skyrocketing price tag comes as the world is roiled by surging electricity demand and as it weathers various supply disruptions, some caused by what appear to be changing weather patterns.

In all, CERA says, the construction of new generating capacity that would have cost $1 billion in 2000 would cost $2.31 billion if construction began today.

According to the index, all types of power plants are feeling the pinch. Components and construction materials for nuclear power plants scored the biggest run-up in costs, up 173% -- nearly tripled -- since 2000. Most of that increase has taken place since 2005. Costs for turbines used to generate wind power more than doubled, at 108%, and natural gas-fueled and coal-fired plants saw their capital costs nearly double, up 92% and 78%, respectively.

I was going over Server Technology's white paper on Planning, Implementing and Application of Cabinet Power Distribution Units, and found a useful set of links for power calculator from Server vendors.

The power ratings found on the product nameplates are notoriously high when compared to its actual power consumption. The reason for this is that manufacturers design power supplies to be used across multiple product lines and also in anticipation of additional features and upgrades in the future. Best practices are to check the equipment manufacturers' website to see if they have real-world power loads and capacities available. See Appendix A for a listing of power calculator tools available from Cisco, Dell, HP, IBM and Sun.

Appendix A

Power Calculation Tools
Cisco
http://tools.cisco.com/cpc/
Dell
http://www.dell.com/calc
Hewlett Packard
http://www.hp.com/go/bladesystem/powercalculator
IBM
http://www.ibm.com/systems/bladecenter/powerconfig
Sun
Power Calculators are available but not in one centralized location. At www.sun.com use their search feature and query for "Power Calculator"

The Dell Web based tool on above site is informative.

Energy efficiency is not just a problem of power (kWh), not just a problem of cooling (BTUs), not just a problem of performance (MIPS) and not just a data center issue — it's all of the above. It is your ultimate challenge. We believe it's time to confront all of these issues all together. To get started, use Dell's Data Center Capacity Planner to better understand your "Green print."

Web Version

• What's new in version 2.0
• DCCP Tutorial
• Start the DCCP 2.75 web version

EFYTimes.com has an article about Green IT survey by Net App India.

Monday, May 26, 2008:  According to a new survey by NetApp, a majority of companies believe that the power utilities should provide an incentive on net savings achieved on power consumption in data centres. Delhi based companies see this as most important, with 77 per cent of respondents wanting 10 per cent or more of the power consumption savings as incentives; Mumbai follows with 66 per cent respondents wanting this incentive; followed by Bengaluru with 62 per cent.

Reduction in energy costs is the key driver of Green IT for 36 per cent of all respondents, according to the finding, and 25 per cent believe that simplification of IT infrastructure is the key driver, while environmental concerns top the list for only 23 per cent of respondents. And, 11 per cent of the respondents gave priority to savings in real estate requirements (to host IT assets at a data centre) as the key driver of Green IT.

The 'Green IT' survey conducted in India also finds 40 per cent respondents believe that server consolidation can significantly reduce power consumption in data centres, while 27 per cent believe that storage efficiency can curtail data centre power consumption. Thirty-two per cent of respondents who recommend server consolidation also believe storage efficiency can have a significant impact on power consumption.

Around 28 per cent of the respondents admit that they have not yet measured power, cooling and space costs in their data centres, whereas 19 per cent say that power, cooling and space eats up over 20 per cent of their IT budgets. Twenty-six per cent put this figure at 10-20 per cent of the IT budget, while 22 per cent respondents put it at 5-10 per cent.

I've seen this graphic many times in presentations, and I wanted more background on it.

The original article is here. With the details behind Christian's calculations.

In a recent article, the 3-year Energy Cost to Acquisition Cost ratio (EAC) [3] was introduced as a metric to understand the cost of energy relative to the cost of the server. Today, for 1U servers this is approaching unity and comes as a surprise to most data center managers. To convince the reader, it is a simple calculation to determine the energy cost of the server:

3-yr Energy Use Cost = 3 yrs x (8760 hrs/yr) x ($0.10/kWhr) x (Server Power in kW) x PUE (1) where PUE is the Power Usage Effectiveness [3] or the Data Center Electrical Load over the IT Electrical Load. For a well managed data center this value is usually about 2.0 (or less), which implies that for every Watt of server power, an additional Watt is consumed by the chillers, UPSs, air handlers, pumps, etc. Indications are that for some data centers this value can be as high as 3.0 [4] and in some cases higher. Usually, this variation is completely due to how well the cooling environment is designed in the data center and has a direct relationship to the energy cost [5].

Using equation (1) for a 1U server (which, when fully configured, costs about $4,000 and consumes about 500 W) and a PUE of 2.0 results in a cost of energy of $2,628. This is almost as much as the server itself, but the reality is that in many cases the cost is much higher. In Japan, energy costs are twice as much, so this number would be double. To make matters worse, in data centers where the cooling design is poor (PUE = 3.0, for example), the cost of energy would be 50% higher.

This means that the energy cost would be $3,942 in the U.S. and $7,884 in Japan. Clearly, there can be huge savings in this energy cost by focusing on optimizing the cooling in the data center as shown in the articles identified earlier [3,5].

Unfortunately, the energy usage is not the only cost driver that scales with power. In 2005, fundamental research [6] was published showing that the infrastructure cost is a big portion of the TCO and quantified the real cost drivers in the data center, which included the amortized cost of the power and cooling infrastructure. This research shows that a fundamental problem is the over-provisioning of cooling due to poor cooling and the lack of understanding of the environment. In addition, The Uptime Institute has also introduced a simplified way for estimating the cost of data center infrastructure [7] based on Tier ratings. For brevity, only Tier IV data centers (with dual redundant power throughout) will be examined since this is the recommended approach for mission critical operations. The Uptime Institute’s Infrastructure Cost (IC) equation for a Tier IV data center is as follows:

IC = Total Power x ($22,000/kW of UPS output for IT) + ($2370/m² of allocated raised floor for IT) (2)

While, admittedly, the authors state that there is a large error band around this equation, it is very useful in capturing the magnitude of infrastructure cost. In fact, it is this author’s contention that this equation could be fine tuned for more accuracy using PUE because poor cooling will mean that more infrastructure will be needed.

However, that discussion is beyond the scope of this paper. Again, looking at the 500 watts of power consumed by the 1U server and using equation (2) and ignoring the IT space the server occupies, the cost of infrastructure to support that server would be enormous at $11,000. The reality is that this cost would be amortized over 10 to 15 years so real annual cost of the infrastructure is $1,100 per year. For the 3-year life of the server, this equates to $3,300 or again close to the cost of the server. Note that there is also an adjustment in the cost as a result of the space occupied by the server, but its calculation is beyond the scope of this discussion.