Searching for a Higgs in a Haystack.
Posted by Science Oxford on May 23, 2011 | comments
Alex Moss reviews “Untangling the Web” at Science Oxford Live on 28 April.
We start amidst the searing 40 degree heat of Morocco, at one of the most lively spots on Earth, the ironically named Djemaa el Fna “Square of the dead”. A place consisting of thousands of tourists, gamblers, midnight ramblers, hecklers, hagglers, conmen, policemen, holy men, business men and those suspiciously resembling men, all bombarding around the square. Just a short moment spent in detached observation, would reveal money pick-pocketed at a rate of 100 of collisions per second, to spend at the Souqs, where a man goes to purchase anything he desires, but will end up buying most things he doesn’t. Amidst the desert dust and haze of confusion, a digital camera, taking 10, 20, 30 photos per minute, behind it the author, frantically clicking, in search for that elusive photo that will promise him all the appraisal, respect, read notifications and “likes” facebook can offer.
Whilst in Switzerland, a slightly less energetic scene unfolds. Amidst the searing heat inside the Large Hadron Collider, at places reaching temperatures 100 000 times hotter than the heart of the Sun, two beams each containing thousands of proton bundles (2,808 to be exact) each containing billions of protons (1.1 x 1011 to be pedantic) are propelled in opposite directions at 99.9999991% of the speed of light. Just one second spent observing this scene, would reveal the protons making a 27 Km round trip 11,000 times, having smashed into one another 600 million times. Amidst this haze and confusion, 6 cathedral sized digital cameras or detectors, taking millions of photos per second, behind them, thousands of physicists, in search of that elusive Higgs Boson, a particle that will unlock the secret to mass and other deep mysteries that the universe has to offer.
As I retreat exhausted to the relative sanctuary inside the walls of my Marrakesh Riad, I have just enough energy remaining to breathe a sigh of despair at the 100s of photos I must now analyse, in the hope of finding a few worthy of Facebook. My sense of impending despair is somewhat tempered by the fact that around the globe, thousands of scientists face the similarly insurmountable task of analysing their own collection of photos, 37 terabytes worth daily (enough to fill more than 1.7 million dual-layer DVDs a year) from the trigger happy detectors at CERN the European Organization for Nuclear Research.
Nestled away in the quiet suburbs of North West Geneva, straddling the the Franco–Swiss border, CERN is operating the world’s largest particle physics laboratory. As much as 175 meters below is The Large Hadron Collider (LHC), the world’s largest and highest energy particle collider. It was constructed by CERN in collaboration with over 10 000 scientists across the globe, with an aim of addressing some of the deepest mysteries of the universe.
The LHC lies in a tunnel 27 kilometres in circumference, a length necessary in order to accelerate beams of protons and lead nuclei towards speeds approaching that of light. The beams of particles are subsequently collided together, with hopes of demonstrating the existence of the hypothesised Higgs Boson, the particle which is theorised to be responsible for the masses of all other particles.
So you see, we’re not so different, me and the quantum physicists at CERN. Whilst they may be burdened or blessed with the task of analysing trillions of proton collisions in the hope of finding a theoretical sub atomic particle, that has never once been observed. I’m faced with the only marginally more daunting task of analysing hundreds of photos, in the hope of finding one of my girlfriend that she’ll permit to be uploaded on Facebook.
So, exactly how does one go about selecting, sifting, sorting and all the other s-verbs, that amount of raw data? According to Dr Ian Collier of the Rutherford Appleton Laboratory in his presentation held at ScienceOxford live last month, the answer lies in Grid computing.
As Dr Collier explained, in order to processes the 15 000 000 petabytes of data created annually, instead of opting for the creation of a centralised metropolis at CERN housing 1000s of physicists, a network or “Grid” of computers was employed. Based in 34 countries around the globe including Canada, France, the USA, Germany, Italy, the Netherlands, Spain, Taiwan and the UK. These satellite centers are connected to the main CERN site by 10 Gigabit fibre-optic links, (around a 1000 times more than your home broadband service delivers and several orders of magnitude more than they promised).
Though the technology may seem advanced, Dr Collins explains that to calculate the masses of data received from the LHC daily, the scientists chose to focus on utilising a large number of relatively inexpensive bits of equipment, as opposed to a select number of ‘state of the art’ supercomputers. A strategy that was in part responsible for the success of companies such as Google.
The system is similar to you or I, employing a league of extremely trusting and non-judgemental friends to sort through our own holiday snapshots. The more friends we employ for this purpose the less of a burden we place on each one, and depending on the closeness of the relationship (and the content of the photos) the less compensation they expect (blackmail) in return.
Dr Collins continues to address another concern, which is best explained in the following terms. Your grandmother has, in a fleeting decision to avoid paying the ferryman for a few more years, created an account on this “face-space” thing she has heard so much about. This seems like a respectable and progressive pastime for you dear old nan to undertake, until you find her desperately forcing her rolls of 35mm film into the CD tray. The technology your grandparents used to store such data, is not the technology we use today. And though you may be sneering at the ignorance of the pre-apple generation, your own grandchildren will be doing the same to you, whilst you struggle to understand why your beloved iPhone refuses to sync with your grandchild’s already outdated holographic quantum computer. And with the rate of technological expansion, you may not have to wait until your twilight years for this scene to occur.
This is in short the problem facing Collins and the other scientists in charge of processing the data at CERN. As the main infrastructure for the LHC computing GRID was conceived in the early 90s, it was fast approaching redundancy, even from it’s inception. For the time being this does not pose a problem, with the data stored away on regular hard drives and then magnetic tape for the long haul. However, after the LHC has reached its shelf-life in around 15 years time, the masses of data will be continued to be re-analysed and studied for generations to come.
Much in the same way that a detective (say Morse as this is Oxford) may revisit the files of one of Jack the Rippers brutal slayings, in hope of cracking the case with modern forensic knowledge, Scientists of the future using the science of their time, may use the data to discover new secrets of the universe not realised during the LHC’s maiden voyage. In 100, 50, 20 or even 10 years, the physicists of the time will be utilising new forms of technology and must be able to gain access to read the vast archives of LHC data. So Collins and his colleagues working with CERN must choose and invest carefully in the storage technology of the future, lest they run the same fate as my mother, gradually coming to terms with the fact that her Beta-Max wedding tape is essentially lost in the annals of time.
Concerning the overall structure, the LHC computing grid is organised into several Tiers. The data stream from the detectors provides approximately 300 GBs per second, which is filtered for “interesting events”, providing a resultant stream of about 300 MBs per second. This data is then sent to Tier 0 which is CERN’s central computing system. The computers at this tier will first process information before subsequently organising it into packets, ready for dispatch via the fibre-optic links to the Tier 1 sights. The roles of the Tier 1 sites, the homestead of Dr Ian Collins and the Rutherford Appleton Laboratory, is to further process data, divide it up and send it on its way to over 160 Tier 2 sites consisting of predominately universities or scientific institutions.
According to Collins, modern scientific research relies on massive computers to unpick the enormous amounts of data that are generated. He continues to explain that one issue encountered with such a large network is ensuring the security of the data. The sheer amount of data that is received from the experiments at the LHC prohibits the usage of firewalls, so in order to prevent unauthorised access, the system employs a process of identification and authorization of those using it.
Particular credit must be given to Collins during the question period of the presentation. In response to a genuinely good-natured and honest but ultimately absurd question, enquiring as to whether or not sub-atomic particles do or do not fall in love, Collins managed to address the questioner without patronising her, nor resorting to intellectual snobbery. In response to the scientific naïvety and ignorance of others, to merely berate or cast them off as foolish, in the author’s opinion, only serves to create enemies of science.
Much like the problems facing the scientists at CERN, the huge amount of data and information produced in today’s society can leave one utterly bewildered as to, which is valuable, which is dangerous, which is useful, which is true. Such a confusion, coupled with the media’s fascination with distorting science, and humankind’s intrinsic conviction ‘that a conspiracy theory is better than no theory at all’, has led to a generation of people, tightly bound to concepts of pseudo-science, misinformation and superstition. In respect to a regular member of the public, the prospects of mining such data for the few golden nuggets of truth are low indeed.
If I may utilise my already worn comparison of myself to the scientists working with CERN: in the same respect that their role is to sort through vast amounts of data, in order to determine what is and isn’t relevant, it is also the role of Science Oxford, and Science communicators in general, to prospect the endless mines of information that saturate the world, in order to educate the public as to what information is valid, meaningful, reliable and true. But science communicators should be wary of assuming an arrogant and pious position as the sole bestowers of truth. We must also aid the public with enough knowledge and understanding so that they are able protect themselves from the endless swathes of pseudo-science, misinformation, outright lies and ignorance that confronts them, and rumour has it that the LHC will soon engulf the universe, so we best not waste time!
What do you think?