I’ve just read Diego’s response to an ill-informed NYT article about data-center power efficiency [1]. This makes me wonder, how much server use does each person have?

Google

Almost everyone uses Google, most of us use it a lot. The main Google product is also probably the most demanding, their search engine.

In a typical day I probably do about 50 to 100 Google searches, that sounds like a lot, but half of them would probably be for one topic that is difficult to find. I don’t think that I do that many Google searches because I generally know what I’m doing and when I find what I need I spend a lot of time reading it. I’m sure that many people do a lot more.

Each Google search takes a few seconds to complete (or maybe more if it’s an image search and I’m on a slow link), but I think it’s safe to assume that more than a few seconds of CPU time are involved. How much work would each Google search take if performed on a single system? Presumably Google uses the RAM of many systems as cache which gives a result more similar to a NUMA system than one regular server working for a longer time so there is no way of asking how long it would take to do a Google search with a single server. But I’m sure that Google has some ratio of servers to the rate of requests coming in, it’s almost certainly a great secret, but we can make some guesses. If the main Google user base comprises people who collectively do an average of 100 searches per day then we can probably guess at the amount of server use required for each search based on the number of servers Google would run. I think it’s safe to assume that Google doesn’t plan to buy one server for every person on the planet and that they want to have users significantly outnumbering servers. So even for core users they should be aiming to have each user only take a fraction of the resources that one server adds to the pool.

So 100 searches probably each take more than 1 second of server use. But they almost certainly take a lot less than 864 seconds (the server use if Google had one server for every 100 daily requests which would imply one server for each of the heavier users). Maybe it takes 10 seconds of server use (CPU, disk, or network – whichever is the bottleneck) to complete one search request. That would mean that if the Google network was at 50% utilisation on average then they would have 86400*.5/10/100 == 43 users per server for the core user base who average 100 daily requests. If there are 80M core users that would be about 2M servers, and then maybe something like another 4M servers for the rest of the world.

So I could be using 1000 seconds of server time per day on Google searches. I also have a Gmail account which probably uses a few seconds for storing email and giving it to Fetchmail, and I have a bunch of Android devices which use Google calendars, play store, etc. The total Google server use on my behalf for everything other than search is probably a rounding error.

But I could be out by an order of magnitude, if it only took 1 second of server use for a Google search then I would be at 100 server seconds per day and Google would only need one server for every 430 users like me.

Google also serves lots of adverts on web sites that I visit, I presume that serving the adverts doesn’t take much resources by Google standards. But accounting for it, paying the people who host content, and detecting fraud probably takes some significant resources.

Other Big Services

There are many people who spend hours per day using services such as Facebook. No matter how I try to estimate the server requirements it’s probably going to be fairly wrong. But I’ll make a guess at a minute of server time per hour. So someone who averages 3 hours of social networking per day (which probably isn’t that uncommon) would be using 180 seconds of server time.

Personal Servers

The server that hosts my blog is reasonably powerful and has two other people as core users. So that could count as 33% of a fairly powerful server in my name. But if we are counting server use per USER then most of the resources of my blog server would be divided among the readers. My blog has about 10,000 people casually reading it through Planet syndication, that could mean that each person who casually reads my blog has 1/30,000 of a server allocated to them for that. Another way of considering it is that 10% of a server (8640 seconds) is covered by me maintaining my blog and writing posts, 20% is for users who visit my blog directly, and 3% is for the users who just see a Planet feed. That would mean that a Planet reader gets 1/330,000 of a server (250ms per day) and someone who reads directly gets 1/50,000 of a server (1.72s per day) as I have about 10,000 people visiting my blog directly in a month.

My mail server which is also shared by a dozen or so people (maybe that counts as 5% of a server for me or 4320 seconds per day). Then there’s the server I use for SE Linux development (including my Play Machine) and a server I use as a DNS secondary and a shell server for various testing and proxying.

Other People’s Servers

If every reader of a Planet instance like Planet Debian and Planet Linux Australia counts as 1/330,000 of a server for their usage of my blog, then how would that count for my own use of blogs? I tend to read blogs written by the type of people who like to run things themselves, so there would be a lot of fairly under-utilised servers that run blogs. Through Planet Debian and Planet Linux Australia I could be reading 100 or more blogs which are run in the same manner as mine, and in a typical day I probably directly visit a dozen blogs that are run in such a manner. This could give me 50 seconds of server time for blog reading.

Home Servers

I have a file server at home which is also a desktop system for my wife. In terms of buying and running systems that doesn’t count as an extra server as she needs to have a desktop system anyway and using bigger disks doesn’t make much difference to the power use (7W is the difference between a RAID-1 server and a single disk desktop system). I also have a PC running as an Internet gateway and firewall.

Running servers at home isn’t making that much of an impact on my computer power use as there is only one dedicated 24*7 server and that is reasonably low power. But having two desktop systems on 24*7 is a significant factor.

Where Power is Used/Wasted

No matter how things are counted or what numbers we make up it seems clear that having a desktop system running 24*7 is the biggest use of power that will be assigned to one person. Making PCs more energy efficient through better hardware design and better OS support for suspending would be the best way of saving energy. Nothing that can be done at the server side can compare.

Running a server that is only really used by three people is a significant waste by the standards of the NYT article. Of course the thing is that Hetzner is really cheap (and I’m not contributing any money) so there isn’t a great incentive to be more efficient in this regard. Even if I allocate some portion of the server use to blog readers then there’s still a significant portion that has to be assigned to me for my choice to not use a managed service. Running a mail server for a small number of users and running a DNS server and a SE Linux development server are all ways of wasting more power. But the vast majority of the population don’t have the skills to run their own server directly, so this sort of use doesn’t affect the average power use for the population.

Nothing else really matters. No matter what Google does in terms of power use it just doesn’t matter when compared to all the desktop systems running 24*7. Small companies may be less efficient, but that will be due to issues of how to share servers among more people and the fact that below a certain limit you can’t save money by using less resources – particularly if you pay people to develop software.

Conclusion

I blame Intel for most of the power waste. Android phones and tablets can do some amazing things, which is hardly surprising as by almost every measure they are more powerful than the desktop systems we were all using 10 years ago and by many measures they beat desktop systems from 5 years ago. The same technology should be available in affordable desktop systems.

I’d like to have a desktop system running Debian based on a multi-core ARM CPU that can drive a monitor at better than FullHD resolution and which uses so little power that it is passively cooled almost all the time. A 64bit ARM system with 8G of RAM a GPU that can decode video (with full Linux driver support) and a fast SSD should compete well enough with typical desktop systems on performance while being quiet, reliable, and energy efficient.

Finally please note that most of this post relies on just making stuff up. I don’t think that this is wrong given the NYT article that started this. I also think that my estimates are good enough to draw some sensible conclusions.

• [1] http://blog.diegodoval.com/2012/09/23/a-lot-of-lead-bullets-a-response-to-the-new-york-times-article-on-data-center-efficiency/

18 months ago when I signed up with Virgin Mobile [1] the data transfer quotas were 200MB on the $29 per month plan and 1.5G on the $39 per month plan. About 4 months ago when I checked the prices the amounts of data had gone up on the same plans (2.25G for $39 per month from memory). Now $39 per month gets only 500MB! It seems that recently Virgin has significantly reduced their value for money.

Virgin does have an option to pay an exgtra $10 per month for 2GB of data which gets you 2GB per month if you sign up for 24 months. That is reasonably good value, when I first signed up with Virgin I paid $39 per month to get extra data transfer, now I could use the $29 plan for phone access and spend $10 per month on data with a Wifi gateway device.

On top of this the phone plans aren’t nearly as good value as they used to be. When I signed up with Virgin the Sony Ericsson Xperia X10 was “free” on a $29 plan, at the time that was a hell of a phone. I believe that the Samsung Galaxy S3 currently occupies a similar market position to the one that the Xperia X10 did 18 months ago – so it shouldn’t be much more expensive. But Virgin are offering the Galaxy S3 for $21 extra per month over 24 months on the $29 plan – a total cost of ($29+$21)*24==$1200 while offering the same amount of calls and data transfer for $19 per month ($19*24==$456) when you don’t get a phone – this makes the price of a Galaxy S3 $1200-$456==$744 while Kogan [2] sells the same phone for $519 + postage!

The cheapest phone that Virgin is offering is a Galaxy S2 for $5 per month on a $29 plan which when compared to $19 per month for the same plan without a phone makes the phone cost ($5+$10)*24==$360. Kogan sells the Galaxy S2 for $399 so there’s a possibility of a Virgin plan saving some money over buying a phone from Kogan. But given the choice of $360 for a Galaxy S2 from Virgin and the Kogan prices of $839 for a Galaxy Note 2, $349 for a Galaxy Nexus, $469 for a Galaxy Note, $529 for a Galaxy S3, and $219 for a HTC One V I find it difficult to imagine that anyone would think that the $360 Galaxy S2 is the best option.

I’ve previously investigated dual-sim phones for cheap calls and data [3] but they didn’t seem like good value at the time because the “free” phones offered by the telcos used to be a good deal. Now it seems that none of the telcos are offering good deals on phones so with my needs the way to go would be to buy a Samsung Galaxy S3 or Samsung Galaxy Note 2 from Kogan, and get the $19 plan from Virgin – probably with a $10 per month extra fee to get an extra 2GB of data. For my wife the best option would be to keep using the Xperia X10 on a $19 per month plan as she doesn’t seem to have any problem with the Xperia X10 that justifies spending hundreds of dollars.

I idly considered getting a portable Wifi-3G device to use a cheap pre-paid 3G data option ($10 per month) and a cheap phone plan without data (maybe $10 per month), but decided that it’s not worth the effort. The Virgin $19 plan gives me free calls to my wife and lots of calls to other numbers (more than I can use) and an extra $10 gives me all the data transfer I need. To use a Wifi-3G device would involve buying such a device and the hassle of carrying it and using it, that wouldn’t save money for at least a year and would be annoying.

The sudden decrease in data quotas is a real concern though. It’s an indication that the telco cartel in Australia is pushing prices up, that’s not a good sign. LTE is nice, but 3G with better quotas would be more generally useful to me.

• [1] http://www.virginmobile.com.au/
• [2] http://www.kogan.com/au/
• [3] http://etbe.coker.com.au/2011/10/27/dual-sim-amaysim-contract/

Generally I haven’t had a positive attitude towards the NBN. It doesn’t seem likely to fulfill the claims of commercial success and would be a really bad thing to privatise anyway. Also it hasn’t seemed to offer any great benefits either. The claim that it will enable lots of new technical developments which we can’t even imagine yet that aren’t possible with 25Mb/s ADSL but which also don’t require more than the 100Mb/s speed of the NBN never convinced me.

But one thing it could really do well is to give better Internet access in remote areas. Ideally with static or near-static IPv6 addresses (because we have already run out of IPv4 addresses). Currently 3G networks do all sorts of nasty NAT things to deal with the lack of IPv4 addresses which causes a lot of needless pain if you have a server connected via 3G. One of the NBN plans is for wireless net access to remote homes, with some sanity among the people designing the network such NBN connections would all have static IPv6 subnets as long as they don’t move.

I’m currently working on a project that involves servers on 3G links. I don’t have a lot of options on implementation due to hardware and software constraints. So if the ISPs using the NBN and the NBN itself (for the wireless part) could just give us all IPv6 static ranges then lots of problems would be solved.

Of course I don’t have high hopes for this. One of the many ways that the NBN has been messed up is in allowing the provision of lower speed connections. As having an ADSL2+ speed NBN connection is the cheapest option a lot of people will choose it. Therefore the organisations providing services will have to do so with the expectation that most NBN customers have ADSL2+ speed and thus they won’t provide services to take advantage of higher speeds.

The Age reports that Adobe has ceased development of Flash for mobile systems [1]. This is described as leading to an improvement in the web experience for iPhone and iPad users, but the more important thing is that it will improve the experience for everyone. The Flash plugin has always been a resource hog and has never been properly supported on all the common platforms. Also most sites that use Flash never needed to as there were other ways of getting equal or better results without it.

Now that Flash is officially on the path to obsolescence everyone can move to HTML5.

I use the following configuration directives in my Squid configuration to block Flash, I selectively enable Flash for the few web sites which use it for useful things. Blocking flash in this manner means that desktop systems which have the Flash plugin installed probably won’t be vulnerable to Flash security flaws as it is unlikely that one of the few sites that I permit to send Flash to my network would end up hosting hostile Flash code.

acl swf url_regex swf$ swf\?
acl swftype req_mime_type -i ^application/x-shockwave-flash$
http_access deny swf
http_access deny swftype

Wikipedia has a comparison of HTML5 and Flash. One interesting benefit that is claimed for Flash is that it allows DRM and it supports inserting commercials and in other ways giving the user an experience that they don’t want. It seems that to put some more nails in the Flash coffin we need tools to suck video from Flash sites regardless of DRM and which skip commercials.

• [1] http://www.theage.com.au/technology/technology-news/adobe-surrenders-to-apple-in-mobile-flash-war-20111110-1n85p.html

Currently Dick Smith is offering two dual-SIM mobile phones for sale in Australia. One is the LG T510 for $99, but it only supports GSM on each SIM. This might be a good phone for someone who needs to receive both work and personal calls and doesn’t want to carry two phones, but the lack of 3G support is a major limit on what can be done with the phone.

The other phone is the Huawei U8520 which supports 3G on one SIM and GSM on the other. It costs $249, runs Android 2.2, has a 320*480 display, and a 3.2 megapixel camera. For comparison the LG Optimus One is a single-SIM phone with similar specs that only costs $179 from TeleChoice, so there is a 40% price premium to pay for a dual-SIM phone.

When I first heard about dual-SIM phones (before they were commonly and cheaply available in Australia) I had thought that it would be a good option for using a cheap 3G broadband SIM along with a SIM for voice calls from one of the cheaper pre-paid mobile companies. But the helpful guy at Dick Smith informed me that Amaysim offers good pre-paid deals for voice and data [1]. With 10G of data quota to be used in one year for $100 and reasonable rates on voice calls it should be easy to keep under $200 per annum if you don’t use many calls.

Rene Cunningham has described how to use a pre-paid data-only plan on the Optus network with VOIP for most outbound calls [2]. To do that he is paying $30 every 6 months to keep his old number for inbound calls for which he gets $30 of credit, with Amaysim you can pay $10 every 3 months to get the same result with $40 per annum call cost instead of $60. As Amaysim are on the Optus network the result should be the same as long as Amaysim have enough capacity for IP data transfer. Rene uses an iPhone but the same result can be achieved with an Android phone.

If by using VOIP the cost of running a phone on Amaysim was reduced to something like $160 per annum (with a possibly optimistic aim of $20 per annum for outbound VOIP calls) then over two years that could save $376 over a $29 per month contract. A Virgin $29 contract includes a Sony Ericsson Xperia X10 which is a fairly nice phone if you can deal with the short battery life and the fact that it’s locked to Android 2.1. An Xperia X10 can be bought on Ebay for less than $376 but the hassle of setting up VOIP and Amaysim will be more effort than it’s worth to save $100 over two years.

A couple of my relatives have phone contracts that are about to expire. I’m not going to set them up on VOIP as it’s too much effort for too little benefit and the dual-SIM phone really isn’t an option. I will recommend Virgin contracts with Xperia X10 phones or Amaysim with their existing phones (2yo smart phones that are still quite usable).

• [1] https://www.amaysim.com.au/mobile-plans/amaysim-as-you-go.html
• [2] http://rene.bz/9-months-on-a-data-only-mobile-phone-plan/

Anne Rettenberg wrote an article for Psychology Today that is critical of the idea of online dating [1], she cites one example of a man who visited a prostitute due to being depressed at his lack of success in online dating to support her claim.

The first big problem with her article is that she doesn’t mention the different experiences that male and female customers presumably have on online dating sites. I don’t know what it’s like for women on the dating sites so I can’t comment on that. But I’m sure that someone who works as a counselor could provide some useful insight into this matter. Also she didn’t even give a mention to the issue of gay/lesbian dating sites.

The next issue is that she didn’t offer any good advice for who should use online dating sites and what their aims should be.

Rejection

In dating in real life (IRL) it’s expected that the man make the first move, and therefore women end up rejecting lots of guys for various reasons. Anne seems to think that rejection online is somehow worse than rejection IRL, it probably is for some people but that certainly isn’t the case for everyone. The way that lots of dating sites seem to work is that women place adverts, men respond to them, and then the women reply to a small subset of the email that they receive. “Rejection” in this case isn’t a matter of telling someone that you aren’t interested, but of merely not replying to their mail.

From my discussions with a few men who’ve used online dating sites the strategy seems to be to send out initial messages to a few dozen women every day and then maybe get a few responses a day. For the messages that get no response you will never know whether the other person found someone else first, wasn’t interested in you for some reason, or just didn’t bother checking their email. The only comparison to IRL rejection is that which happens after phone numbers have been exchanged, which isn’t going to be that common (and has the same issues regardless of how the people met).

Guys, relax about the women who don’t respond to your first message. They probably get 100+ responses to their advert and don’t have time to even read half of them. If you get rejected later in the process then you can look through your email archive at a later date to try and discover what went wrong.

The Aim of Online Dating

The fact that Anne’s client visited a prostitute suggests that maybe he wasn’t really after a relationship. In which case using one of the many online services for finding sex partners might have been a better option.

Generally it seems that a good strategy is to try and have fun. I don’t know any men who have married someone from a dating site, but the general opinion seems to be that they are still worth using. If you meet someone in a bar then you might end up having a drunken conversation that is drowned out by loud music. If you meet someone over the Internet then you can have a quiet conversation over the phone – which seems to be a better way to get to know someone (and generally more pleasant for anyone who’s not an alcoholic). I think that men who have no immediate aim other than finding someone nice to talk to will do better than those who aim to score quickly.

Of the men I know who married women they met over the Internet (but not through online dating sites) I wonder how many of them would have ended up married if they hadn’t used the online dating sites first. It seems that men who regularly communicate with women outside formal situations (work etc) will have a better chance of impressing someone that they like than those who lack such experience.

Introverts

There are a lot of people who really can’t function in a bar. With the way our society works it seems that anyone who can’t handle the bar scene really should consider online dating.

How Counselors can Help

It seems to me that someone who is seeing a counselor and who is considering a new way of finding a SO should ask their counselor for advice first. It also seems to me that a good option might be to ask their counselor for advice in online dating. Instead of being unsuccessful and depressed a man who was seeing a female counselor could do well to ask her advice for how to impress women on the Internet. This is probably a business opportunity for female counselors who can advise men on such things, among other things it seems that seeing an “online relationship coach” would be perceived in a more positive manner than seeing a counselor or psychologist for the more traditional reasons.

• [1] http://www.psychologytoday.com/blog/beyond-don-juan/201107/can-online-dating-make-you-depressed

Periodically Free Software people from other countries visit Melbourne on business trips. Usually when someone is sent any distance on a business trip (IE to Australia from anywhere other than NZ) they will stay in a good hotel (4 star or better), this generally means that they have a pool in their hotel. 5 star hotels and the newer serviced apartment hotels tend to have really good pools (1/4 olympic size isn’t uncommon). Hotel pools are very under-utilised, their main purpose AFAIK is to boost the hotel star rating – my experience is that it’s not common to meet other people in a hotel pool.

While food and drinks are often banned in the pool area my observation is that the only rule which matters is “no glass”. So hotel pools are almost ideal for pool parties, you just need to drink from cans or from drinks in plastic bottles that are poured into plastic cups.

If any Free Software person finds themself staying in a hotel in Melbourne Australia with not much to do in the evenings or weekends then one option is to call for a pool party. I’ve asked on a local mailing list and it seems that there is enough interest for a small party, the local mailing lists can be used to arrange a party.

Also one thing to note is that some hotels have outdoor pools, while Australia is generally a warm place the southern parts of Australia (such as Melbourne) get quite cold in winter, an outdoor pool is not going to be fun for the colder half of the year. So getting a hotel with an indoor pool is very important during the April-September period.

Virgin Mobile Excessive Billing

In my previous post about the Xperia X10 I mentioned being billed excessively for bandwidth use [1].

On Friday when I phoned Virgin I was repeatedly told that there was nothing that could be done about the bill. I demanded to be transferred to someone with authority to change my bill, I was told that wasn’t possible so I demanded to be transferred to a supervisor. As no supervisors were available at the time I had to be called back.

The “Supervisor”

Tonight I received a call from someone who’s employed to deal with disputed bills, after explaining the situation I was offered a discount of $150, that’s a 64% discount on the $234. I rejected that offer and explained that I only used 7% more data than was permitted in the month and only did so because of a previous phone call with a Virgin representative telling me that the billing period ended at the month end (which I couldn’t verify at the time because the Virgin CRM system was down [2]). Thus it doesn’t seem reasonable to charge me 7* the regular billing amount for the entire month.

I was asked what I thought would be a reasonable extra charge, I suggested 7.4% more than the $39 contract, IE an extra $3 – that wasn’t well received. Then I suggested that as the $49 “Smart Cap” plan would have covered the extra data use it could be reasonable for me to pay $49 for the month. The Virgin representative then offered me a $200 discount on the bill and I accepted that. So I got a 85.5% discount on the disputed amount, that’s not too bad, it’s enough to make this scam less profitable to Virgin (particularly when I publish information about it) and enough to make it not worth the effort of a TIO complaint.

That makes it a bill of $102.36 instead of the original $302.36. That is $29 for one month of the “Big Cap $29” for my wife, $39 for one month of the “Smart Cap $39” for me, $33.99 for Virgin excessive bandwidth charges (down from $233.99), and $0.37 “Account Charges” which is described as a “Payment Processing Fee” – presumably a fee for paying by credit card.

The Ethical Issue

To make it clear, I think that only a minority of mobile phone customers have an ability to track their bandwidth and understand their bill which equals mine. I also think that only a minority of customers are as willing as I am to argue with random people about such things. I am sure that employees of Virgin Mobile know this and they devised a billing plan to trap customers. I think that this doesn’t comply with the ethical standards that Richard Branson has advocated in several TV interviews that I have watched.

The Next Issue

Now the only remaining issue is that I signed up for an offer “3 Months Free ^ access fee on Topless, Smart & Easy caps with this phone. Save up to $267, just by buying online” which continued “Promo code: reggae Use this promo code & we’ll automatically credit your first 3 months’ access fees“, but so far they have not credited it. It’s a pity I didn’t realise this before and get it done in the same call.

Conclusion

I’m starting to regret not keeping my LG U990 Viewty with a pre-paid SIM from Lebara [3] or Amaysim [4]. Amaysim has a minimum pre-paid value of $10 which lasts for 90 days, and the SIM will then receive calls for 180 days after the credit has run out – this means $10 per 260 days (an average of $7.11 per year) if you just want to receive calls. Lebara has a period of 120 days before the number expires, so that means it costs $10 per 210 days or an average of $5.75 per year to receive calls only.

Telstra charges $150 for one year of mobile net access which allows 10G of data transfer [5]. So it seems that a good option would be an old phone and a new Android tablet. $150 per year ($12.50 per month) for mobile net access and something between $5.75 and $40 per year for phone access (depending on whether you want to make calls) gives enough savings over the cost of a post-paid plan to buy a decent tablet.

I think that the Android phone is definitely the best deal for my wife because she really only wants to carry one device, also she likes having longer conversations on the phone so the plan where she gets up to 450 minutes for $29 per month is a reasonable deal – particularly as calls to me are free and calls to her relatives will be free if they get Virgin phones (something that now seems doubtful given their over-billing).

I’m happy to carry multiple devices (my past record is having four mobile phones and getting a tailor-made coat to fit all the geek stuff I carried) so a phone and a tablet is a reasonable option for me. But the Xperia X10 will do the job for the next two years and I’ve got no great regrets.

On the up-side, Virgin did send me a couple of free movie tickets after their CRM stuff-up.

• [1] http://etbe.coker.com.au/2011/04/17/more-about-xperia-x10/
• [2] http://etbe.coker.com.au/2011/02/23/virgin-mobile-crm-upgrade-failure/
• [3] http://www.lebara-mobile.com.au/
• [4] http://www.amaysim.com.au/
• [5] http://www.telstra.com.au/bigpond-internet/mobile-broadband/prepaid/rates/index.htm