The power of Mozilla = .2mW/user


We talk a lot about Data.  It’s a fundamental component of Mozilla’s 2010 Goals, second only to Mozilla’s role as a centerpiece of the Internet.  In September, Mitchell talked about “usage data.”

IT, and especially Operations, collects of a lot of Usage Data. Most of this Usage Data is trended and used in capacity planning and debugging.  It’s useful internally and probably not that interesting externally.

A lot of the more interesting data comes in the form of bandwidth graphs, NetFlow analysis, DNS statisticsload balancer and router performance graphs, China bandwidth and China GSLB traffic distubution.

We’ve recently begun trending power usage from Mozilla’s primary data center in San Jose1.  The following graph shows wattage per row2:

San Jose power consumption

The genesis of this was inspired by Nobel Laureate Al Gore. I had the opportunity to see Mr. Gore give the keynote at the RSA Conference in San Francisco last April. His comment, “Make the invisible, visible,” resonated and stuck with me.

On average we’re using about 50kW (and remember, that’s a point-in-time measurement, the amount of power we’re using at any given time). I find that number to be too abstract — I’ll try to put that number into more meaningful terms (this calculator was helpful):

  1. If the local utility were to bill us for this, we’d get a bill for about 36MWh.  For perspective, my PG&E bill for last month was for 484kWh.
  2. The Tesla Roadster could go about 370km (230 miles) with 50kWh.
  3. In California, 1 kWh of electricity generates 0.878707 lbs of CO2 emissions.  Our San Jose data center generates  43.94 lbs of CO2 (or 0.02 metric tons).  For comparison, an average car (20 mpg) generates 44 lbs of CO2 after driving about 45 miles.3
  4. According to the DOE, the average U.S. home uses approximately 936kWh of electricity per month, which is about 31 kWh/day.  Daily we consume 1.2MWh.  At that rate we could power ~38 homes every day.
  5. Or, with roughly 250 million Firefox users, 50kW is 200 milliwattsmicrowatts/user (.2mW) (or 144Wh/month).

That last one is an interesting metric to measure power usage.  It’s more a measure of efficiency, efficiency per user.

I’m left with some questions:

  1. Who else is showing this sort of energy usage data?
  2. Is this a useful metric?  Can I use this to compare with others, apples-to-apples?  Does it mean anything?
  3. Is Mozilla’s data center energy usage good?  Bad?
  4. How can we use this data to help drive business decisions?
  5. Does energy consumption change during Firefox releases?

Lastly, I want to point out that the graph on this page is a link to a live image.  The data for this graph is pulled every five minutes and the graph image is updated twice an hour.  At some point this blog post will get lost in my archives so I’ve created a Wattage page with this graph and more historical graphs.  As we’re able to, I’ll add data from Mozilla’s other data centers.


1 Admittedly, this graph under reports the true energy impact. It doesn’t take into account the amount of energy required to cool the data center space we’re in and is obviously missing both the Amsterdam and Beijing data centers.

2 This per-row graph shows something interesting. 16/101 has 9 racks and 212 devices in inventory. 16/102 has 6 racks, two of which are just patch panels. All the routing and load balancing hardware is in this row and inventory shows 59 devices. 16/103 has only 7 racks and 231, the most dense row and from the graph, less power hungry than 16/101. It’s also the row with all 141 Mac Minis.

3Calculations provided by Mira @ TerraPass.


Categories: Mozilla

9 responses

  1. morgamic wrote on :

    Great post, Matthew. I think we should try to run more metrics like this so we can explain to other companies how much CO2 emissions you can save by optimizing the front-end of websites. Perceived speed, quicker-loading pages and saved bandwidth don’t speak as loudly as CO2 saved and/or direct dollars and cents saved.

    So, say we cut the # of HTTP requests by 50% — what amount of savings would that incur at Mozilla?

  2. Archaeopteryx wrote on :

    You could
    1. cook 44827 cups of coffee or tea per day with that (continental European style, sea level, 0,25 l cup, 8°C cold water) – more than a cup each 2 seconds.
    2. let 24131 people call other 24131 people with theirs iPhone 3Gs

  3. Axel Hecht wrote on :

    Interesting data for sure. As a european, 20kg are the best metric unit, opposed to howmany whichever pounds or centi-tons.

    As you noted that cooling is excluded, we should get a ballpark number for that at least. I learned that heat contributes significantly to the power consumption of the system at large. And can make actions like bying more efficient hardware a good idea.

  4. mrz wrote on :

    @Axel: Re: buying more efficient hardware, we’ve made a conscious decision to standardize on low-voltage CPUs. Last year we started moving away from single 1u rack mount servers towards blades (and LV Xeons).

  5. Wil Clouser wrote on :

    Graphs, numbers, and interesting data. Nice work. :)

  6. Tristan wrote on :

    Excellent work, mrz! “make the invisible visible”, then we’ll work on reducing it.

    Limiting energy consumption is urgent, and I’m totally thrilled to see Mozilla move in the right direction. Keep up the good work!


  7. John Slater wrote on :


  8. Brian King wrote on :

    Next up: Float the Mozilla data center out in the Pacific. Even better, the Atlantic is windier and likely has stronger currents. Get your goggles on!

  9. open4energy wrote on :


    I believe you have opened the door to a key issue for all open source projects.

    How do they differentiate from commercial software when it comes to energy management.

    If, as we believe, the open nature of open source results in cleaner code, produced under the critical eye of our peers, how will this translate into energy savings.

    But it goes further, how will open source companies operate their own IT systems to save energy.

    And, what software will be provided by the open source community to support energy savings.

    My appreciation to Cacti and RRDTool who made this data collection possible.

    See and