Category Archives: Physics

More cores, less speed

Posted on by

An interesting set of simulations at Sandia of multi-core systems have been reported:

A team of researchers simulated key algorithms for deriving knowledge from large data sets. The simulations show a significant increase in speed going from two to four multicores, but an insignificant increase from four to eight multicores. Exceeding eight multicores causes a decrease in speed. Sixteen multicores perform barely as well as two, and after that, a steep decline is registered as more cores are added.

The reason for this is fairly well known, but it’s nice to see numbers put to the effect..

The problem is the lack of memory bandwidth as well as contention between processors over the memory bus available to each processor.

The original Sandia press release has more information.

A Tale of Two Transport Hacks

Posted on by

In the USA a court has ordered that three MIT students not talk at DEFCON about their security assessment of the Massachusetts Bay Transit Authority (MBTA) fare cards. Apparently the court believes that “discussing the flaws at a public conference constituted a ‘transmission’ of a computer program that could harm the fare collection system“, which is pretty sad. There are more documents at Cryptome on the case. Their presentation was to include a cryptanalysis of the Mifare “Classic” card, which takes us to our second case..

Bruce Schneier reports that a group of Dutch researchers have won in court to be able to publish their own cryptanalysis of that very same Mifare Classic card, with the court stating:

Damage to NXP is not the result of the publication of the article but of the production and sale of a chip that appears to have shortcomings.

An outbreak of common sense that the MIT students could only dream of. I wonder if they could appeal and cite this case as grounds to have the judgement overturned ?

Dr Brian May, guitarist

Posted on by

Congrats to Brian May on getting his PhD published!

Indeed, when I looked at the Springer Astronomy homepage, I was greeted by the incongruous sight of Brian May next to a snippet about the thesis – directly above the blurbs for the newly-published Biographical Encyclopedia of Astronomers and Dictionary of Minor Planet Names & Addendum.

Oh, and no, it’s not this Brian May, he got his PhD some time ago.. 🙂

CERN Large Hadron Collider – Big Science

Posted on by

Thanks to Jeremy for pointing out to me a photo feature on the LHC at CERN. Some really brilliant pictures of truly Big Science getting built.

View of the CMS detector at the end of 2007 (Maximillien Brice, (c) CERN)

View of the CMS detector at the end of 2007 (Maximillien Brice, (c) CERN)

Of course, if you’re worried about this causing the world to end it’s probably worth reading the info they put out in 2003, which simply points out that the Earth will have seen this (and much more) before:

The Large Hadron Collider (LHC) can achieve an energy that no other particle accelerators have reached before, but Nature routinely produces higher energies in cosmic-ray collisions. […] Whatever the LHC will do, Nature has already done many times over during the lifetime of the Earth and other astronomical bodies.

To put things into proportion:

Speculations about microscopic black holes at the LHC refer to particles produced in the collisions of pairs of protons, each of which has an energy comparable to that of a mosquito in flight.

So you can put your brown paper bags away now.. 🙂

CSIRO & BOM report – “Drought: Exceptional Circumstances” (not)

Posted on by

For those looking for the joint assessment by the Bureau of Meteorology and the CSIRO that’s all over the news at the moment, you can find it on the MAFF website. There is also a web page listed for the data and analysis in the report, but it’s not working yet (I guess they forgot the webmaster doesn’t work on weekends).

It’s about 35 pages long and is fairly technical, but not overly daunting. The content, however, is pretty scary. For us in Victoria is it predicting:

  • by 2010-2040, exceptionally hot years are likely to affect about 75% of the region, and occur every 1.3 years on average;
  • by 2010-2040, exceptionally low rainfall years are likely to affect about 10% of the region and occur about once every 12 years on average;
  • by 2030, exceptionally low soil moisture years are likely to affect about 11% of the region and occur about once every 9 years on average.

Historically it says that Victoria and Tasmania are down 109 mm in rainfall since 1950 and average temperature is up by almost 0.8C over the same 50 year timescale.

The most worrying thing is that these predictions are based on a lower level of CO2e emissions than we are currently tracking towards.

Observations since 1990 show that we are tracking the highest IPCC emission scenario, called A1F1, but climate simulations have not been performed using the A1FI scenario. Most climate research institutes around the world did simulations using the mid-range emission scenarios, called A1B and A2. Hence, in this report, projections for the next 20 to 30 years are based on simulations using mid-range emission scenarios.

So if we carry on how we’re doing now, then the reality could be much worse..

Help Search for the Missing 1999 Mars Polar Lander

Posted on by

The Planetary Societys Emily Lakdawalla has blogged about an interesting project up on their website at the moment, trying to rope in volunteers to help NASA locate Mars Polar Lander using images from the HiRISE camera on the Mars Reconnaissance Orbiter. Emily writes:

What I would really love is if any of you readers out there who wanted to join in the search would write to me and let me know which image you’re searching, or ask me to assign you one, so that we can spread out the effort of all the volunteer searchers and make sure each image is examined by multiple people. I’ve also given some guidelines on how to report anything that you think might be a piece of the missing Mars Polar Lander. So if you want to join in the search, go check out that page.

Currently there are 18 images to search through, and the full resolution JPEG 2000 images are over 1GB a shot..

Quote for the day

Posted on by

In 1969 Bob Wilson (later the first director of Fermilab) was called before a hearing of the US Congressional Joint Committee on Atomic Energy to answer questions about particle accelerators. In it Senator John Pastore demanded to know how such a device improved the security of America and Bob Wilsons response of “nothing at all” didn’t go down to well, and so he was prodded further.

His obituary from Cornell in January 2000 puts it like this:

“It has only to do,” Wilson told the lawmakers, “with the respect with which we regard one another, the dignity of men, our love of culture. It has to do with: Are we good painters, good sculptors, great poets? I mean all the things we really venerate in our country and are patriotic about. It has nothing to do directly with defending our country except to make it worth defending.”

I have to concur.