Slashdot Mirror

All this wasted energy sounds like an argument for some stiff taxes on energy.
  I honestly see no other way to discourage people from wasting so much energy on these things.

Who gets to decide what's "good" use of energy, and what's bad? These guys have been running their computers for 16 years for a contest to break an algorithm that nobody really uses, the keyspace problem is well understood, the message isn't very interesting, and they don't even have enough computing power to actually crack the thing!

I think that's a pretty big waste of energy, but people keep it up anyway. Why should anyone get decide which energy usage they like and which they don't?

I see no 404. http://distributed.net/ try again.

There is already existing infrastructure and projects where people can donate their system's computational power ...

Distributed.net

I had the idea several years ago of doing a computer animated movie using a voluntary distributed computing model, like distributed.net or SETI@Home. Once it was scripted and storyboarded and the animation plan was complete, individual frames (or even portions or layers of frames) would be farmed out to folks to run on their computers. The complicated part would be that folks would almost certainly be able to assemble parts of the film prior to release, but that might be OK. Multiple scenes could be produced, so the plot could have multiple possible endings. If I were to produce part of the film on my home machine, I'd want to go to the theater and see how it all works.

My script idea was of three or four young kids, who fly in virtual fighter planes over a landscape that is based on the real structure of the net as a geographic metaphor. Of course cities and other facilities in the virtual landscape would match up fairly well with the real landscape, since the net does that already. I think the father of one of the kids is the one who invented the visual metaphor immersion system. They would discover one guy who was a bully or hood at school IRL flying his own plane (as a result of hacking into the father's system and stealing an early copy, that has some flaw the bad guy doesn't know about), and dropping 'bombs' (metaphorical visualization of inserting hacks) onto websites around the net. (I came up with the idea when most hacking was recreational, not commercial or political.) So these kids would have to use their own skills and tools to fight the hacker guy in virtual space, and also deal with him and his gang in real life.

IMHO that would be a great movie, and I'll bet a large number of slashdotters would love to participate. Heck, the profits could even go to support open source. And now it's possible to do it in 3D.

I even toyed with the idea of setting up a website where folks could work together to build the script and the story board. This could be the first 'open source movie'. I do own meatspace.us ... hmmm.

Plus, there's already a project for cracking encryption and other mathematical challenges using bovine power.

14,000 CPUs haven't cracked RC5-72 in over 2 600 days. http://distributed.net/

I've got it! We can create our own open source network lines. Each person will go to the hardware store and buy 10 meters of fibreoptic cable and dig a trench in front of their house. We can take our spare parts and combine them and make servers! Power to the people! Stickin it to the man! Yeah!!!

You could do something like that, like Fon tried to do some time back in the US and parts of Europe with WiFi (whatever happened to that? I wound up reformatting my router and using standard images because their OS was so unstable).

Or you could do the spare parts thing with Grid / Distributed Computing.

Just sayin'.

The encryption technology used to prevent eavesdropping in GSM (Global System for Mobile communications), the world's most widely used cellphone system, has more security holes than Swiss cheese, according to an expert who plans to poke a big hole of his own.

I hope I am not the first to say: "giggity."

Each GSM phone has its own secret key, which is known by the network. Every time a call is initiated, a new session key for that particular call is derived from the secret key and used to encrypt the call. Nohl aims to crack the session key.

This actually doesn't sound like a bad encryption scheme.

To speed up computing time, the project relies on some components not always found in your standard PC, such as Nvidia Corp.'s CUDA (Compute Unified Device Architecture) graphics cards and Xilinx Virtex field-programmable gate arrays (FPGAs).

So, are those of us without fancy video cards or FPGAs allowed to help? Even if we can't compute keys as quickly?

That's the point Nohl hopes to drive home: The A5/1 algorithm is a broken 64-bit encryption technology, a relic of the Cold War era, when laws prohibited the export of strong encryption technology from the United States. It needs to be replaced--ideally by the much stronger, 128-bit A5/3 system

So GSM itself isn't that insecure, it's that they're using a short key length. This is rather old news then. All they are doing is brute-forcing the whole key space rather than breaking the algorithm. This is basically what brought down RC5-56 and DES (although DES had other flaws as well).

I mis-read my notes; that 100k/s figure for your standard desktop is actually 100M/s and comes from the password cracking competition at distributed.net. According to their current live stats, the fastest single-CPU system (an Intel Core i7 2666Mhz) is cracking ogrng at 204M/s and the average is 5.5M (with a wild standard deviation of 8.6M) and from current live multi-CPU stats, a 4-CPU Intel Core 2 quad-core (16 cores) at 3110MHz is cracking rc572 at 450.8M/s and the average is 36M (stdev=51M). That puts 100M/s at more than a standard deviation above average for even a multi-CPU system and more than ten standard deviations above the average single-CPU system.

The PS3s at 200k apiece look pretty measly now, falling well under the average desktop on Dnet (5.5M). Since even an AMD K6 can crunch away at 300k/s on rc572, it's probably reasonable to say that they're cracking something tougher than anything at Dnet. Generously pinning the PS3 to the Intel Core 2 Quad 3GHz (40M/s) means dividing my Dnet numbers by 200 or multiplying the government's numbers by 200.

At 40M/s times the 60 PS3s, we'd come to 2.4G/s, which can break an 8-character alphanumeric password in a day and an 8-character random printable (includes punctuation et al, 6.5 bits of complexity) in 22.7 days. Bring that to ten characters or six characters plus two words and you're suddenly talking about 500 years. Assuming they actively upgrade with no loss to data (to fit Moore's Law) and you're looking at 9 years ( log2(500) ).

I figure military-grade is probably 10-100G/s (with continuous upgrades according to Moore's Law), which would still take 3-7 years to find a 10-char password but blows through the 8-char password in 4-7 hours.

I mis-read my notes; that 100k/s figure for your standard desktop is actually 100M/s and comes from the password cracking competition at distributed.net. According to their current live stats, the fastest single-CPU system (an Intel Core i7 2666Mhz) is cracking ogrng at 204M/s and the average is 5.5M (with a wild standard deviation of 8.6M) and from current live multi-CPU stats, a 4-CPU Intel Core 2 quad-core (16 cores) at 3110MHz is cracking rc572 at 450.8M/s and the average is 36M (stdev=51M). That puts 100M/s at more than a standard deviation above average for even a multi-CPU system and more than ten standard deviations above the average single-CPU system.

The PS3s at 200k apiece look pretty measly now, falling well under the average desktop on Dnet (5.5M). Since even an AMD K6 can crunch away at 300k/s on rc572, it's probably reasonable to say that they're cracking something tougher than anything at Dnet. Generously pinning the PS3 to the Intel Core 2 Quad 3GHz (40M/s) means dividing my Dnet numbers by 200 or multiplying the government's numbers by 200.

At 40M/s times the 60 PS3s, we'd come to 2.4G/s, which can break an 8-character alphanumeric password in a day and an 8-character random printable (includes punctuation et al, 6.5 bits of complexity) in 22.7 days. Bring that to ten characters or six characters plus two words and you're suddenly talking about 500 years. Assuming they actively upgrade with no loss to data (to fit Moore's Law) and you're looking at 9 years ( log2(500) ).

I figure military-grade is probably 10-100G/s (with continuous upgrades according to Moore's Law), which would still take 3-7 years to find a 10-char password but blows through the 8-char password in 4-7 hours.

I mis-read my notes; that 100k/s figure for your standard desktop is actually 100M/s and comes from the password cracking competition at distributed.net. According to their current live stats, the fastest single-CPU system (an Intel Core i7 2666Mhz) is cracking ogrng at 204M/s and the average is 5.5M (with a wild standard deviation of 8.6M) and from current live multi-CPU stats, a 4-CPU Intel Core 2 quad-core (16 cores) at 3110MHz is cracking rc572 at 450.8M/s and the average is 36M (stdev=51M). That puts 100M/s at more than a standard deviation above average for even a multi-CPU system and more than ten standard deviations above the average single-CPU system.

The PS3s at 200k apiece look pretty measly now, falling well under the average desktop on Dnet (5.5M). Since even an AMD K6 can crunch away at 300k/s on rc572, it's probably reasonable to say that they're cracking something tougher than anything at Dnet. Generously pinning the PS3 to the Intel Core 2 Quad 3GHz (40M/s) means dividing my Dnet numbers by 200 or multiplying the government's numbers by 200.

At 40M/s times the 60 PS3s, we'd come to 2.4G/s, which can break an 8-character alphanumeric password in a day and an 8-character random printable (includes punctuation et al, 6.5 bits of complexity) in 22.7 days. Bring that to ten characters or six characters plus two words and you're suddenly talking about 500 years. Assuming they actively upgrade with no loss to data (to fit Moore's Law) and you're looking at 9 years ( log2(500) ).

I figure military-grade is probably 10-100G/s (with continuous upgrades according to Moore's Law), which would still take 3-7 years to find a 10-char password but blows through the 8-char password in 4-7 hours.

I mis-read my notes; that 100k/s figure for your standard desktop is actually 100M/s and comes from the password cracking competition at distributed.net. According to their current live stats, the fastest single-CPU system (an Intel Core i7 2666Mhz) is cracking ogrng at 204M/s and the average is 5.5M (with a wild standard deviation of 8.6M) and from current live multi-CPU stats, a 4-CPU Intel Core 2 quad-core (16 cores) at 3110MHz is cracking rc572 at 450.8M/s and the average is 36M (stdev=51M). That puts 100M/s at more than a standard deviation above average for even a multi-CPU system and more than ten standard deviations above the average single-CPU system.

The PS3s at 200k apiece look pretty measly now, falling well under the average desktop on Dnet (5.5M). Since even an AMD K6 can crunch away at 300k/s on rc572, it's probably reasonable to say that they're cracking something tougher than anything at Dnet. Generously pinning the PS3 to the Intel Core 2 Quad 3GHz (40M/s) means dividing my Dnet numbers by 200 or multiplying the government's numbers by 200.

At 40M/s times the 60 PS3s, we'd come to 2.4G/s, which can break an 8-character alphanumeric password in a day and an 8-character random printable (includes punctuation et al, 6.5 bits of complexity) in 22.7 days. Bring that to ten characters or six characters plus two words and you're suddenly talking about 500 years. Assuming they actively upgrade with no loss to data (to fit Moore's Law) and you're looking at 9 years ( log2(500) ).

I figure military-grade is probably 10-100G/s (with continuous upgrades according to Moore's Law), which would still take 3-7 years to find a 10-char password but blows through the 8-char password in 4-7 hours.

I mis-read my notes; that 100k/s figure for your standard desktop is actually 100M/s and comes from the password cracking competition at distributed.net. According to their current live stats, the fastest single-CPU system (an Intel Core i7 2666Mhz) is cracking ogrng at 204M/s and the average is 5.5M (with a wild standard deviation of 8.6M) and from current live multi-CPU stats, a 4-CPU Intel Core 2 quad-core (16 cores) at 3110MHz is cracking rc572 at 450.8M/s and the average is 36M (stdev=51M). That puts 100M/s at more than a standard deviation above average for even a multi-CPU system and more than ten standard deviations above the average single-CPU system.

The PS3s at 200k apiece look pretty measly now, falling well under the average desktop on Dnet (5.5M). Since even an AMD K6 can crunch away at 300k/s on rc572, it's probably reasonable to say that they're cracking something tougher than anything at Dnet. Generously pinning the PS3 to the Intel Core 2 Quad 3GHz (40M/s) means dividing my Dnet numbers by 200 or multiplying the government's numbers by 200.

At 40M/s times the 60 PS3s, we'd come to 2.4G/s, which can break an 8-character alphanumeric password in a day and an 8-character random printable (includes punctuation et al, 6.5 bits of complexity) in 22.7 days. Bring that to ten characters or six characters plus two words and you're suddenly talking about 500 years. Assuming they actively upgrade with no loss to data (to fit Moore's Law) and you're looking at 9 years ( log2(500) ).

I figure military-grade is probably 10-100G/s (with continuous upgrades according to Moore's Law), which would still take 3-7 years to find a 10-char password but blows through the 8-char password in 4-7 hours.

distributed.net was doing about 35 billion keys per second in 1998.

You are an idiot

Their next project is OGR-26, as they announced today.

Pfff, 25-mark. Wake me up when they get the 26-mark. </unimpressedslashdotuser>

Instead of sleeping, why not help them find it?

They have just started OGR-NG which will search for 26-mark and higher-order rulers. For now you will have to use a prerelease client.

http://n0cgi.distributed.net/cgi/dnet-finger.cgi?user=bovine

http://www.distributed.net/download/prerelease.php

They have just started OGR-NG which will search for 26-mark and higher-order rulers. For now you will have to use a prerelease client.

http://n0cgi.distributed.net/cgi/dnet-finger.cgi?user=bovine

http://www.distributed.net/download/prerelease.php

This says what a Golomb ruler is.

http://www.distributed.net/ogr/

Yet nothing in this article or link says "why should we care what it is?" Who uses Golomb rulers? What are they used for?

Ah no, that would be http://www.distributed.net/

-Iceman

Go have a look at distributed.net to see how effective thousands of computers are at breaking RC5 encryption with a 72 bit key. So far it's been going for 5.5 years or so and is at 0.495% of the keyspace http://stats.distributed.net/projects.php?project_id=8
This is only a 72 bit key by brute force. Try breaking AES 256, I dare you.

    Just for reference:

  http://www.distributed.net/history.php
  January 28, 1997
        RC5-32/12/7 (56-bit) Secret Key Challenge begins

Patent filed Jan 7, 1999. Prior art: http://www.distributed.net/ January 28, 1997: RC5-32/12/7 (56-bit) Secret Key Challenge begins

Case closed.

There have been 16 different viruses/worms/trojans that have installed the distributed.net client. http://www.distributed.net/trojans.php

I'm sure other distributed computing projects have seen the same thing happen to them.

BOINC was a huge step forward for distributed computing, and yes, it was originally designed for use with SETI@Home.

However, in all fairness, distributed.net was really one of the first to successfully pull off the model of using volunteer's spare CPU cycles. I was using DNETC back in 1997, SETI@Home was released in 1999, and BOINC didn't show up until (I believe) 2003.

BOINC was a significant incremental improvement over the widely used clients that came before it (in that it could readily support third-party projects), but it was far from the first successful example of distributed-computing for the masses.

It already happened with distributed.net. They have quite a long history of such problems: http://www.distributed.net/trojans.php/

Here's what I did for a client. I installed the distributed.net client on all their machines with a different ID per. If one went missing, I just waited until it started posting again.

One was recovered. I don't know out of how many thefts, but it worked.

http://stats.distributed.net/participant/psearch.p hp?project_id=5&st=coreyfro

Some are still posting to this day.

216264(-1) K6231862@coreyfro.com 13-Feb-2001 08-Feb-2002 361 791
218871(-1) K3342513@coreyfro.com 31-Jan-2001 20-Jun-2002 506 729
219222(-1) K4151626@coreyfro.com 29-Jan-2001 18-Jul-2001 171 721
223856(-2) K5557748@coreyfro.com 08-Feb-2001 02-Jul-2002 510 622
223908(-2) K2863155@coreyfro.com 29-Jan-2001 21-Oct-2001 266 621
224051(-2) K3456175@coreyfro.com 20-Jan-2001 31-Dec-2001 346 618
224360(-2) K4553312@coreyfro.com 22-Jan-2001 10-Jun-2002 505 612
225611(-3) K6211864@coreyfro.com 27-Mar-2001 09-Aug-2001 136 588
227645(-5) K8631173@coreyfro.com 17-Aug-2001 30-Jun-2002 318 549

I've seen distributed.net's processing client flagged as spyware or a virus on a few occassions.

Plus other distributed projects, like the ones from http://www.distributed.net/

"SETI pioneered the idea" ???

http://www.distributed.net/ was doing it long befor seti@home

You're mistaken if you think SETI@home was the distributed computing pioneer. The distributed.net encryption cracking project started in 1997. I ran it from '97-'99 or so through the RC5 and DES contests.

http://www.distributed.net/history.php

Which came first, SETI@Home or distributed.net?

I know dnetc's been running on at least one of my machines since 1997 or thereabouts.

As a sysadmin, this sounds neat -- but I haven't seen any computing environments that need that kind of horsepower yet. But, I can't wait to crank-up my distributed.net ranking.

At my last contract, we used IBM Bladecenters -- Linux in a dev/QA environment, and they had prolly the largest load-generator farm I've ever seen. It wasn't the CPUs that were maxed, tho -- just the network.

It's well known that foolproof remote verification of client code from a server is impossible, so confidence in XBox Live's ability to detect this mod seems misplaced at best.

That'll be sure to get the team stats up!

Now, who do I call for a free demo?

Can you run the most recent distributed.net client for this system, and provide us with the typical keys/second value you obtain?

While not a perfect benchmark by any means, and highly dependent upon which core is used (ie. the algorithm and implementation), it is often a useful test to quantify how a machine performs.

I use to play with this idea 4-5 years ago. A small team was going to look into building FPGA PCI boards that could be used with http://www.distributed.net/ to help crack DES/RC5/*insert-your-choice-encryption-here*.

Folding@Home? Nah, if you look at distributed.net's credit page you'll see me (grub) a couple of times. So I have loyalty to them :) (although it's been ages since I did anything for d.net...)

Mr. Beberg,

In the interest of full disclosure, you might have pointed out that you left distributed.net in not so friendly terms with the rest of the team. Don't sweat, so did I (I joined distributed.net as a core coder after you left, so we never met.) But whenever I mention something that might be construed as negative about distributed.net in public, I try to disclaim my potential biases. Hence I'm doing this favor to you and the rest of Slashdot readers.

It wouldn't hurt to mention that distributed.net is no longer about key cracking only -- even if you have a grudge against RC5-72 (I also think it's fairly pointless at this moment in time), OGR is a completely valid project.

Actually, SETI@Home runs like http://www.distributed.net./

it's another contest hosted by RSA... it's like $10k again, nothing major... basically people do it for the stats (it proves your technical manhood by having your name on the top 100 or whatever) I think most people understand that the contest being solved in an inevitability over time, and that their chances of actually winning money are extremely small, but it can be fun to race your friends in stats.

For those interested in this sort of thing, http://www.distributed.net/ runs like SETI@Home - lots of small individual clients working together to brute force encryption keys.

Yeah, but SETI@Home is searching for intelligent life.

Distributed is searching for things that probably never should have been encrypted in the first place.

When I was in the military, about 90 percent of the SECRET level documents should have been declassified to RESTRICTED or CONFIDENTIAL, and as one of the few subject matter experts, I spent a lot of time doing that because noone else could read the things that got overclassified.

I mean, you can encrypt your golf scores. And some people do. Security is usually spent on the wrong areas, and not often enough on the areas that should be protected or encrypted. Just ask New Orleans what they think about mismanagement of resources ...

For those interested in this sort of thing, http://www.distributed.net/ runs like SETI@Home - lots of small individual clients working together to brute force encryption keys.

> Actually, reading on, it looks like the author
> really doesn't have a clue.

Indeed. He says:

> This Diffie-Hellman key exchange protocol is
> used, for example, by RSA and PGP.

RSA and DH key exchange are two different animals. The formulas involved have a certain similarity but the principles are different. RSA is an asymmetric encryption scheme based on the hard problem of factoring, and provides authentication. DH key exchange is a "key exchange" protocol based on the hard problem of discrete logs, and is non-authenticating (meaning it is vulnerable to a man in the middle attack).

Also, the article states:
> By the late 1990s, security researchers had been
> able to break the 56-bit DES algorithms in as
> little as three hours.

But the 1998 break by EFF in combination with distributed.net took over 23 hours. Details, details... see

http://www.distributed.net/des/

This article is an embarrasment to IBM. They have people who know better!

"Face it the market is only in it for FAST x86, nobody cares about power. And if they did they want to see it from intel, or AMD."

Really? I bought an Athlon XP 2500 333 because there was a motherboard for $10 after rebate. Coughed up the dough for the memory too, plus I had to buy a video card because it's 1.5VAGP - they got me in the end, spent well over 200 that started out as 10 bux, still content. In any case, I was still happily chugging along on my 300A Celeron clocked to 450, still working like a charm after 6 years, used mostly for this, browsing the net, reading, playing games. Even MS Rise of Nations Trial was fast on it because it has an nVidia Geforce2MX, that's what counts. And this Athlon does not feel 5x fast.. at all.. maybe twice - the program start time is still mostly controlled by the disk speed, with its 10 ms access time/7200 rpm that hasn't changed much (capacity though went from 10 Gig to 120 Gigs in 6 years). Maybe if I was an http://distributed.net/ RC5 freak trying to brute force crack keys, or if I did heavy video compression, I may need the juice. But this new CPU cooks.. while I read Slashdot.. the room warms up in the winter, I can feel it, unlike with the good old celery. I for one would be quite content to buy a chip that's cheap, consumes 5W and gives 1.5GHz equivalent performance.

Fastest I've personally witnessed is 38Mkeys/s on a dual G5, 18Mk/s on a dual opteron.

Anyone from IBM care to visit http://www.distributed.net/download/clients.php and download/build/run and report the results?

X86/Win32 -- 73%
X86/Linux -- 11%
PowerPC/Mac OS X -- 11%

The remaining 5% is divided among dozens of other combinations.
http://stats.distributed.net/misc/platformlist.php ?project_id=8&view=tco

I have to note that the PowerPC client for distributed.net is very good, a single 1.2 GHz G4 performs on par with a dual 2.4 GHz P4. So, these statistics suggest that ~5.5% of the CPUs is running Mac OS X.