|
|
Subscribe / Log in / New account

Some impressions from Linaro Connect

By Jonathan Corbet
March 13, 2013
One need only have a quick look at the LWN conference coverage index to understand that our community does not lack for opportunities to get together. A relatively recent addition to the list of Linux-related conferences is the series of "Linaro Connect" events. Recently, your editor was finally able to attend one of these gatherings: Linaro Connect Asia in Hong Kong. Various talks of interest have been covered in separate articles; this article will focus on the event itself.

Linaro is an industry consortium dedicated to improving the functionality and performance of Linux on the ARM processor; its list of members includes many of the companies working in this area. Quite a bit of engineering work is done under the Linaro banner, to the point that it was the source of 4.6% of the changes going into the 3.8 kernel. A lot of Linaro's developers are employed by member companies and assigned to Linaro, but the number of developers employed by Linaro directly has been growing steadily. All told, there are hundreds of people whose work is related to Linaro in some way.

Given that those people work for a lot of different companies and are spread across the world, it makes sense that they would all want to get together on occasion. That is the purpose of the Linaro Connect events. These conferences are open to any interested attendee, but they are focused on Linaro employees and assignees who otherwise would almost never see each other. The result is that, in some ways, Linaro Connect resembles an internal corporate get-together more than a traditional Linux conference.

So, for example, the opening session was delivered by George Grey, Linaro's CEO; he used it to update attendees on recent developments in the Linaro organization. The Linaro Enterprise Group (LEG) was announced last November; at this point there [George Grey] are 25 engineers working with LEG and 14 member companies. More recently, the Linaro Networking Group was announced as an initiative to support the use of ARM processors in networking equipment. This group has 12 member companies, two of which have yet to decloak and identify themselves.

Life is good in the ARM world, George said; some 8.7 billion ARM chips were shipped in 2012. There are many opportunities for expansion, not the least of which is the data center. He pointed out that, in the US, data centers are responsible for 2.2% of all energy use; ARM provides the opportunity to reduce power costs considerably. The "Internet of things" is also a natural opportunity for ARM, though it brings its own challenges, not the least of which is security: George noted that he really does not want his heart rate to be broadcast to the world as a whole. And, he said, the upcoming 64-bit ARMv8 architecture is "going to change everything."

The event resembled a company meeting in other ways; for example, one of the talks on the first day was an orientation for new employees and assignees. Others were mentoring sessions aimed at helping developers learn how to get code merged upstream. One of the sessions on the final day was for the handing out of awards to the people who have done the most to push Linaro's objectives forward. And a large part of the schedule (every afternoon, essentially) was dedicated to hacking sessions aimed at the solution of specific problems. It was, in summary, a focused, task-oriented gathering meant to help Linaro meet its goals.

There were also traditional talk sessions, though the hope was for them to be highly interactive and task-focused as well. Your editor was amused to hear the standard complaint of conference organizers everywhere: despite their attempts to set up and facilitate discussions, more and more of the sessions seem to be turning into lecture-style presentations with one person [Hong
Kong] talking at the audience. That said, your editor's overall impression was of an event with about 350 focused developers doing their best to get a lot of useful work done.

If there is a complaint to be made about Linaro Connect, it would be that the event, like much in the mobile and embedded communities, is its own world with limited connections to the broader community. Its sessions offered help on how to work with upstream; your editor, in his talk, suggested that Linaro's developers might want to work harder to be the upstream. ARM architecture maintainer Russell King was recently heard to complain about Linaro Connect, saying that it works outside the community and that "It can be viewed as corporate takeover of open source". It is doubtful that many see Linaro in that light; indeed, even Russell might not really view things in such a harsh way. But Linaro Connect does feel just a little bit isolated from the development community as a whole.

In any case, that is a relatively minor quibble. It is clear that the ARM community would like to be less isolated, and Linaro, through its strong focus on getting code upstream, is helping to make that happen. Contributions from the mobile and embedded communities have been steadily increasing for the last few years, to the point that they now make up a significant fraction of the changes going into the kernel. That can be expected to increase further as ARM developers become more confident in their ability to work with the core kernel, and as ARM processors move into new roles. Chances are, in a few years, we'll have a large set of recently established kernel developers, and that quite a few of them will have gotten their start at events like Linaro Connect.

[Your editor would like to thank Linaro for travel assistance to attend this event.]

Index entries for this article
ConferenceLinaro Connect/2013


to post comments

Some impressions from Linaro Connect

Posted Mar 15, 2013 0:10 UTC (Fri) by giraffedata (guest, #1954) [Link] (8 responses)

in the US, data centers are responsible for 2.2% of all energy use;

This is a classic meaningless ratio, which I've heard a lot recently.

Why not divide by the total amount of resources of all kinds (converting to dollars as common units)? Or by just the electrical energy used by industry? (Rhetorical question - those wouldn't be useful either).

Reminds me of someone who once boasted that a ranch was twice the size of Rhode Island. Someone responded, "that's not saying much. Rhode Island is very small. You know how small? It's half the size of some ranches."

Some impressions from Linaro Connect

Posted Mar 15, 2013 14:31 UTC (Fri) by wookey (guest, #5501) [Link] (7 responses)

Why is it 'meaningless'? It gives you some idea of the scale of (US) data centre energy use in comparison to all other energy use. It might be useful to know if the US-data-centres:US-energy-use and worldwide-data-centre:worldwide-energy-use were particularly different figures, but it's still a useful number. And it shows why it's a good idea to put datacentres in places where the electricity supply is relatively low-carbon. Not as important as putting aluminium smelters there, but still worth doing if you can.

Some impressions from Linaro Connect

Posted Mar 15, 2013 18:07 UTC (Fri) by copsewood (subscriber, #199) [Link]

It's also worth locating in places where the waste heat which needs to be extracted from datacentres could be used in district heating i.e. in cold cities surrounded by suitably high density homes and offices. Unfortunately, without a zero loss electric grid connecting optimal carbon-free electricity generation sources with optimal consumption places, these 2 requirements conflict. As it is, these grid losses tend to be relatively smaller than the energy efficiency advantage of being able to use waste heat for a secondary heating application.

Some impressions from Linaro Connect

Posted Mar 15, 2013 18:38 UTC (Fri) by giraffedata (guest, #1954) [Link] (5 responses)

It gives you some idea of the scale of (US) data centre energy use in comparison to all other energy use

That's what I'm saying is useless. At least for anything related to the various places I see the statistic quoted, such as this talk.

And it shows why it's a good idea to put datacentres in places where the electricity supply is relatively low-carbon.

I don't believe it does show that. What would show that is the cost of putting a typical data center near such an electricity source vs the cost of the extra carbon emitted when that data center is near some other electricity source. Or, starting with just energy numbers, it would be useful to look at the number of megawatts data centers use. But dividing that number by the amount of energy everything else uses doesn't help.

Not as important as putting aluminium smelters there, but still worth doing if you can.

That's the really ironic thing about quoting this particular statistic as justification for working on low-power data center — It's a tiny number. The only place I can see this ratio being useful is if you're responsible for the overall energy policy of the US (or any similar country or the world). You have limited resources, so have to choose on which energy uses to focus. I would say since 97.8% of the energy is used elsewhere, you should not focus on data centers now.

If you're not in the position of having to choose, then you should work on supplying data centers with green electricity sources even if you're only talking about a millionth of the energy used in country. Even a single ton of CO2 removed from the air helps.

Some impressions from Linaro Connect

Posted Mar 15, 2013 20:06 UTC (Fri) by tbird20d (subscriber, #1901) [Link] (1 responses)

That's the really ironic thing about quoting this particular statistic as justification for working on low-power data center — It's a tiny number.
I think you've got it backwards. One imagines that the US energy budget is quite large. Quoting the number as a percentage of that budget gives an indication of how big data center usage is. An isolated number, as opposed to a ratio, wouldn't give that impression.
If you're not in the position of having to choose, then you should work on supplying data centers with green electricity sources even if you're only talking about a millionth of the energy used in country. Even a single ton of CO2 removed from the air helps.
First you argue the number is small, then that even small numbers should be reduced. Which is it? The ratio is only pointing out the that number is big - probably bigger than people think. I'm not sure what your point is relative to that.

Some impressions from Linaro Connect

Posted Mar 16, 2013 2:38 UTC (Sat) by giraffedata (guest, #1954) [Link]

First you argue the number is small, then that even small numbers should be reduced.

Did you notice I switched conditions? The number is small if you're using it to choose which national energy use to look at managing. The number is irrelevant (doesn't matter how small) if you're in a position where the only energy you have to manage is data centers.

The latter people are the ones involved when I've seen the 2.2% number quoted, for example people plannning the future of ARM. So I'm saying they shouldn't quote it. They should use other numbers to decide whether to change data center energy policy or leave it alone.

I mentioned the national energy policy people only in the course of trying to stretch the problem into something where the ratio of data center usage to all usage might matter.

The ratio is only pointing out the that number is big - probably bigger than people think

I know, and I'm saying that's specious. It's common, but wrong, to pick a denominator that makes an emotional impact but is not meaningfully related to the numerator. That's why I originally suggested several other denominators that give quite different numbers but are no less meaningful. And so what if the total energy used by data centers is large? Divide that by the number of data centers, or number of servers, or number of users, and it gets small again. To be useful statistics, ratios have to be between things that vary together in some meaningful way.

Some impressions from Linaro Connect

Posted Mar 16, 2013 1:33 UTC (Sat) by wookey (guest, #5501) [Link]

Like Tim, I don't really follow your logic but here are some plain numbers instead:

Worldwide server, cloud and router energy use: 45GW or 400 TWh/yr (published 2011). Worldwide aluminium smelting energy use (2011): 555TWh/yr (63GW average).

So to a first approximation data centres and aluminium smelting (famously a very energy-intensive process) are in fact quite similar-sized issues in terms of use of supply.

Numbers randomly picked off the internet: http://www.world-aluminium.org/statistics/primary-alumini...

http://www.cs.berkeley.edu/~jtma/papers/emergy-hotnets201...

Some impressions from Linaro Connect

Posted Mar 30, 2013 12:19 UTC (Sat) by Duncan (guest, #6647) [Link] (1 responses)

> That's the really ironic thing about quoting this particular statistic as justification for working on low-power data center — It's a tiny number.

I was struck by yet another bit of missing information: The claim is 2.2% of "energy use", but how is that "energy use" measured -- what does it include and what might it exclude.

The immediate question in my head was "OK, electrical energy only, or all energy?" Does the figure include all powered vehicle use, or just electrical energy? One thing that differentiates the US from many western societies is just how many cars there ARE here, the vast majority still gasoline (or occasionally diesel) powered.

If the figure is only 2.2% of /electrical/ energy, it's a lot less impressive than if it's 2.2% of all artificial energy.

And then there's the whole artificial/natural energy question. Living in Arizona, I appreciate how much energy the sun adds to the equation, for instance, energy that is "used" by plants especially, but also consider the water cycle's use of energy, the energy that creates (and one might argue is used by) our weather, etc. As a purely practical matter, I seriously doubt all /this/ energy was included, or that 2.2% figure begins to look rather impressive indeed!

Which is my point. From my perspective, the 2.2% figure isn't meaningless as a ratio, it's meaningless because the terms of the ratio simply aren't properly defined. It's worse than a suppressed-zero graph, in terms of the potential for "making the numbers lie". Yet we're given literally no idea of just how expansive the definition of "energy use" we're comparing against actually was. 2.2% is interesting as a figure of only electricity energy use, but becomes much MORE so if it's a percentage measure of all artificially generated energy use (again, cars; the other day I was sitting at an intersection here in Phoenix waiting for a bus and watching the cars go by, not really rush hour or anything, and I thought about that happening say 16 hours a day, across all the thousands of intersections here in Phoenix alone... but now I'm thinking about multiplying that by N cities of this side nationwide, then adding all the rural traffic... it'd be interesting to see how /that/ compares to the 2.2% data-center figure...), and even *MORE* so if that 2.2 percent is really of *ALL* energy use as claimed but I seriously doubt as intended, including the energy dissipated by hurricanes, the energy "used" by the sun to heat the pavement that's so hot here in Phoenix in the summer...

It's a scope problem, and without at least /some/ indication of the scope included in that 2.2% claim, it really *IS* meaningless. After all, data-center energy usage could be 100%... if the scope is the measure of all /data-center/ usage.

Does it matter that data centers use 2.2% of energy?

Posted Mar 30, 2013 20:48 UTC (Sat) by giraffedata (guest, #1954) [Link]

Though I thought it was clear what the denominator in the 2.2% figure is (all artificial energy), I don't think it matters. Every one of the possibilities is meaningless for the purposes for which this number is being offered.

I mentioned before that the figure would be useful to a person responsible for US energy policy in deciding where to focus limited resources (which is not what it has been offered for in any place I've seen it). If the denominator is instead of all electrical energy, it would be useful to someone responsible for managing the capacity of the US electrical grid or some piece of it. If the denominator is all energy usage, then it's hard to imagine any use for it at all. Cosmology?

The purpose for which I have seen the figure offered is to argue to computer engineers or makers that it is worthwhile for them to invest in making data centers more energy efficient.


Copyright © 2013, Eklektix, Inc.
This article may be redistributed under the terms of the Creative Commons CC BY-SA 4.0 license
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds