pert

I can't seem to reply to wizard_2's post (I guess it's a designed limit) so here's the post which spawned the encrypted password discussion:

http://news.ycombinator.com/item?id=595231

I think you're very lucky if you're not seeing any contention at peak times. In highly populated areas (central Edinburgh and Livingston in my personal experience), there is huge congestion on the local loop at peak times.

I've not got any bandwidth test results to show this but a friend of mine in Edinburgh does log ping times:

http://cress.dom0.org/cgi-bin/smokeping.cgi?target=DSL

The contention at evenings and weekends is very obvious from these graphs.

They claim it's just to test the technology, but I'm sure a few PR people will have been involved in the announcement and will not be rubbing their hands with glee at the press coverage it's generated.

I found the post to be interesting because it goes into a bit of detail about how local end of the cable networks work.

This also has some relevance to this post:

The Internet sky really is falling (networkworld.com)

http://news.ycombinator.com/item?id=597699

Some ISPs have done very well in staying ahead of the demand curve. Others, such as BT and Virgin Media here in the UK, are not investing in areas of their network where demand has already outstripped supply.

I don't agree with the "lack of access capacity" being a blocker for Internet growth/development. This problem can be solved easily by more investment from the cable/Telecom companies.

Here's a link to what, I guess, is the same video that works from the UK:

http://www.latenightwithjimmyfallon.com/blogs/2009/04/vintag...

I agree. Standard UPSes convert AC to DC and back, in order to power the server PSUs and these conversions are inefficient. If the battery is in the server (after the PSU), you can miss out these conversions and save a lot of power.

A post at the 'Green Data Center Blog' (http://www.greenm3.com/2009/04/close-up-an-old-google-server...) says that the pictures are of a "Google Server which is ready for retirement".

Anyone got a link to the "video tour Google presented of its data center containers"?

They did actually publish their results. It's an interesting read:

http://research.google.com/archive/disk_failures.pdf

Readers of this article might also be interested in Fink: a package manager for OS X, based on dpkg and apt from Debian.

About Fink: http://www.finkproject.org/about.php

Guide to building your own packages: http://www.finkproject.org/doc/quick-start-pkg/index.php

This is the first article that I've read on 'cloud computing' that seemed to be based on fact and not on fancy. Great link! Thanks.

Open source software needs more high-profile success stories like this. Great link. Thanks.

I agree, but this does not guarantee there will be no driver issues and monitoring packages (such as the HP PSP and Dell 'OpenManage') do not come as part of third party operating systems so take time to install. In my opinion this does not detract from Linux in the enterprise, but does mean that it's not quite in the same class as OS X on Apple or [Open]Solaris on Sun.

Although the Linux distributions have come a long way in terms of hardware compatibility, I don't think any of them could match a single-vendor (OS and hardware) solution. I've not used OS X as/on a server, but I assume the hardware monitoring will be well integrated into the hardware and driver issues will be rare. I'd therefore say that it's unfair to compare OS X on the Xserve with the likes of RHEL on Dell, and that a comparison with Sun would be more realistic.

I hadn't read that OS X had become 'real UNIX' (trying to do a 'The Wire' reference there). That's a very smart move by Apple, as it will help them to attract OS X ports of traditional UNIX software, such as the big databases, and will certainly help establish them on the server market.

This article seems a bit confused about how the worm is spread: It is a simple bruit force password attack, targeted at mipsel Linux-based routers.

The post at DroneBL (linked to by the article) seems to be a better read:

http://www.dronebl.org/blog/8

I don't understand how this works. Any box that I ssh to with '-R' just listens on 'localhost'.

The idea is that, instead of using cold water to mix with your boiler water, you use cold water that has been slightly heated by the heat-exchanger. This allows you to use less water from your boiler.