Posts by Steven Jones 1526 publicly visible posts • joined 21 May 2007
If a bunch of spanner-wielders want to use something else, then fine, but the SI units of meter, second, second, ampere etc. and the derived quantities, like the Newton, were agreed on by the scientific community in 1960. They are defined by physicists and it isn't going to change.
The CGS units are not SI units by the way. It may be an alternative metric unit, but SI it is not.
It's the wrong format to my mind. My company VPN system uses a keyfob with a keyboard and a PIN to get you the one time password with a single line LCD. I'm not sure what the point of a colour LCD is (yet more advertising), but I'm pretty sure all the required components could be put into a keyfob format as a 4mm thick credit card is just too big.
However, can I make a plea - if we are going down the electronic identity device line, can the various bodies involved all make sure we only have to carry one? It's bad enough having half-a-dozen 0.5mm think cards for stores, debit, credit and so on. If we had to carry the same number of electronic devices then it would cause chaos.
ps. a small, and oh-so-cheap way of detecting credit card fraud that the banks might want to implement. Just give us the option of receiving an SMS message whenever there is an attempted approval of a payment. That way people will very rapidly find out if their card has been cloned or is using you card number. Just how difficult can it be?
Any pretence that there is any security at all over where a vehicle is registered is shot to pieces by this sort of stuff (which is why, of course, the so many TV programs blurr the numberplates of celebrities and other individuals they feature. Given what mus be tend, if not hundreds of thousands of people that can access this sort of information, then it's almost becoming public data.
This sort of breach is far more important than the cost to the company he worked for. I don't suppose anybody much cares for the privacy of bus companies, but it could as easily have been personal transport.
I also wonder what other sort of personal information the government is prepare to sell to those with a "legitemate interest"?
I have 32bit Windows 7 on a work laptop, the same on a personal laptop and Windows 7 64bit on a home quad-core desktop. All three machines had XP on before, and are all a couple of years old, or a bit more. I've yet to have a single system crash on any of them in about 8 months of total (adding up all three machines). Fortunately I've found drives for all my peripherals.
In essence, the stability and time to boot and responsiveness is a huge improvement and I don't think it is down to just clean builds, but we'll see as the patches build up. Don't have less than 2GB or 2 cores and there's a few changes for changes sake, but basically it is all good. On the basis that Vista was a beta, Windows 7 for, the first time, feels like I could live with it for ever.
OK - so you clearly admit zero latency isn't possible. I get heartily sick of overblown hype about physical impossibilities wowing the credulous. The reason this matters is that over and over again I come across senior managers who swallow this sort of stuff without any conception of what is physically possible.
Of course the latencies are accumulared, but if you are suffering most of the latency on the links and adapter, then you are into the law of diminishing returns by looking only at the switch.
Where things are better, then fine. Tell us what the realistic numbers are, but vendors are always issuing unachievable figures, but when it comes down to hyping up physical impossibilities, then I smell a rat. This stuff matters - a lot.
Another thing - this industry is all about precision. Computers and systems don't work on fluffy sales speak. They work through what in other areas might be called pedantry, but which in hard-core IT is engineering precision.
This guy most know something about the speed of light that Einstein didn't. Unless he's managed to construct space/time worm holes in a data centre, then he's going to have to put up with the same 300 metres per microsecond propagation speed as the rest of creation. Of course that's if he's found a way to route photons through a vacuum. Signals via copper or fibre don't go faster than about 200 metres per microsecond, which means that 50 metre link is never going to have a round-trip time less than 0.5 microseconds, whatever you do with switched, adapters and the like.
There are very low latency interconnects of course, but they all have very short reach for this reason. Indeed the propagation speeds inside computers are a fundamentally limiting factor, let alone on external comms links.
I have something very similar with a 256GB SSD for the system plus some of the more volatile work areas. The rest I have on a pair of 2TB drives setup as a RAID-0 pair. The RAID-0 gives slightly more than double the number of read IOPs as a single disk (although only for multi-threads - any single thread will not see anything faster than a single disk). The other big thing was moving to Windows 7 and the machine just flies along - close down and reboot to fully ready is less than 80 seconds, 30 seconds of which is down to the BIOS. This is despite a very large number of apps.
The hard drives are find for video, large photos, music but you want your meta-data on the SSD. I run LightRoom with the catalogue on SSD and the photos on RAID1 and it's pretty near perfect.
Of course this is hardly a cheap set-up, but fortunately it's the sort of thing you can grow to and, at a push, an 64GB SSD is enough.
How about this for a better idea. You put a 256GB SSD into your PC where you get the benefit all the time and you back it up to an external HDD with a 2.5" drive. Doing it the other way round is, frankly, bonkers. Maybe if you are super-rich you will have SSDs in both, but in that case, do yourself a favour and do it with eSATA which is going to be a whole lot better than USB with it's big driver and protocol overheads.
"done something popular while she was in office... and abolished the British television licence fee?"
Because it simply wouldn't have been popular. We would have ended up with a diet of advertising-funded cheap (in all senses of the word) TV supplemented by very expensive subscription financed TV also riddled with adverisment and made up of largely imported US TV and sports.
Basically what you get now from ITV, C5 and C4 plus Murdochvision. News would follow the rubbish Sky-News or, worse, Fox model. Whatever the problems of the BBC, it is relatively cheap and it has sustained a level of home drama production which Murdoch's crew have singularly failed to do preferring the lazy and risk-averse way of simply buying into US productions plus major sports rights. In the days before advertising fragemented, C4 & ITV could make produce some good stuff. No longer. In that the UK TV business makes money intenationally it's from the selling of inane format TV. It makes people like Simon Cowell very rich, but is not exactly enriching culture.
I would like to see more diversity in publicly-funded broadcasting, so I'm not one of those who think the licence fee (really a hypothecated and regressive tax) should all go to the BBC but, on balance, I think the great majority of the population would wish to retain public funding.
There is one very positive outcome of the loss of deep coal mining in this country, and that it has udoubtedly saved a lot of lives. That's for two reasons. Firstly deep mining is inherently dangerous. Whatever precautions are taken, there will inevitably be some accidents. However, even accidents could be eliminated, there are the health risks inherent in the industry itself. There are thousands of miners whose life has been cut short through working in the filthy atmosphere of coal mines.
it is certainly possible to greatly reduce these two factors, but not eliminate it - at least not until completely automated deep-mining using machinery alone becomes possible.
So 10KW delivered to the IT equipment and only 80W on cooling, lighting, UPS, ventilation, transmission losses, inverters etc.
Forgive me if I'm a trifle sceptical I'd want to know exactly where the measurment points are for this and what lies downstream from them.
However, efficiency it so be applauded. As many runner of large data centres will know, it';s not merely the cost of power that's the problem, it's getting supplied. Power companies often have some limits on the amount they are prepared to deliver to a given location (or at least without paying unimaginably large bills for uplifting distibution networks).
What you describe is called steganography, and anything as crude as using the LSBs is readily detectable. GCHQ would just love it if that was what terrorist cells used. hat said, there are tools to do it properly such that it cannot be so easily detected, although it's not easy. However, you can combine it with encryption as well.
Indeed you might argue that the Truecrypt second level hidden partition encryption is actually a type of steganography in that it hides data in, apparently, areas randomised as part of secure data deletion process
Stating that you;'ve forgotten the password is not necessarily sufficient. This guy made exactly that claim in court, but the jury didn't believe it. Where it can be shown that you've previously had access to the password and it can be shown, beyond reasonable doubt, that you still know it, then you can be found guilty. As to what reasonable doubt is? Well, that's up to the jury to decide.
"The first 50 characters from a favourite book"
I'd make it a bit more complicated than that. There is a finite number of books and it's easily a small enough number to try many variations on the characters at the start of a book as a standard cracking tool. You might want to try combinations from different books, or maybe at different starting points.
The burden of proof for there being an inner encrypted volume would be on the prosecution. They would have to show beyond reasonable doubt in a court of law that such a volume existed which, if you are competent, can't be done using technical means alone. There would have to be evidence, maybe related to personal behaviour or some such which would stand up in court. That will be difficult.
However, the "denier" would have to be very careful indeed as there are plenty of ways that the existence of such an encrypted partition and it's activity can be detected unless the right precautions are taken,
Another thing to note is that who knows if it will be possible to break some encryption in the future using new techniques. As far as I'm aware, with the sole exception of quantum and true "one time" encryption, then there are not mathematical proofs that these schemes can't be broken using some future mathematical principle. If that happens, then decrypted copies of old data might come back to haunt a few people.
You can only be arrested if their is reasonable suspicion of committing an offence so being able to remember or not at that time is not really relevant except if you were able to provide an aliby which the police could check.
In the event, then if it did come to court and still stated you could not remember then it will be up to either the jury or judge/magistrate (depending on the circumstances) to decide on the credibility of this lapse of memory. If the 18th was three days ago then that's going to be a very different thing to the previous "right to silence". Of course there is also the "right to silence" where the rules have changed. Now if you don't answer questions during the interrogation that can be taken into account.
Note that there was never a "right to silence" in court if you were being cross-examined. That right could only be exercised if you weren't called to testify by the defense (the prosecution had no right to call the defendent to testify - only to cross-examination if he/she did appear as a defense witness).
If you couldn't recall a password for an encrypted file and pleaded that you had forgotten it, then it would be up to the prosection to prove beyond reasonable doubt that this was not the case. If you had a whole-disk encryption and other evidence was such that the machine had been in use the previous day, then the amnesia claim might not hold water. However, if you had a three year old file that was encrupted and hadn't been accessed since, then that would be a different thing.
David Allen Green AKA Jack of Kent is going to cover this on his next blog. He's a well known liberal, yet sees there is an obvious conflict of human rights here. The first is the right to privacy against that of handling serious crime; terrorism, paedophilia and the like. As usual, the normal libertarian fringe element sees this in a singularly one dimensional way.
Now this is not to say that there are real issues which could lead to a miscarriage of justice such as forgotten passwords. Then there is the whole issue of "plausible deniability".
Incidentally, failing to reveal a password is not the only case where you can find yourself in legal difficulties for not providing information. For example, if you give somebody permission to drive your car and an offence is committed, and you don't reveal the name then you can be in difficulties. If you don't reveal information demanded by a court order then that will also get you into the realm of criminal law.
Mine is one of the 15% with an HD tuner. In a year's time it will become the norm once the cost difference between standard Freeview & HD Freeview tuners disappears as it surely will.
However, one thing worth noting is the broadcast quality. Freeview HD is not broadcast at a high enough bit rate to get the full benefit in my experience. Maybe once the new bands are worked out, transmission power is upped (limited at the moment in many cases to avoid inter-regional interference) they will be able to up the bit rate as SNR improves, but I rather suspect not. More likely they'll cram more sub-channels into the same space.
High def Freeview recorders are still rather too expensive with limited model choice and there, costs are likely to remain higher as you need bigger disks and more processing capacity.
I too am a bit concerned about buying DRM format books tied into one hardware vendor. However, at £109 if you read a log of "classic" books then it will easily pay for itself. There's a huge amount of stuff out of copyright, although in the UK that isn't until 70 years after the death of the author (so you'll have to wait until 2016 for any HG Well, although you'll find online versions in the US where the copyright period isn't so long).
At this price it gets to the point where you don't have that much to lose.
The £6 per year figure is clearly a nonsense if this is £32bn of extra capital spend. Even using a modest financing cost of 4% per year (and it's probably the allowed ROI number that is most appropriate), £36bn would cost £1.4bn a year just on interest before any capital pay back. Divided among 26m households, then that's about £55 per hear. Put a more realistic financing figure in that, allow for maintenance costs on the extra infrastructure and you might double that or more.
Of course business customers will have to be paying their share too, so it won't all be on householders, but the £6 figure is a joke. Maybe a cummulative extra £6 per year, every year for the next decade and a bit.
it would be nice tho think that some of this capital expenditure will go towards efficiency (both energy and operational costs). Most businesses, when assessing investments, also look at that factor. As it is, regulatory regimes often just motivate companies to increase their capital expenditure as they are simply able to make money on this using the allowed ROI rules.
It shouldn't be forgotten that Norman Wisdom won a BAFTA in a TV drama called "Going Gently" playing a terminally ill cancer patient. Judy Dench played a nurse, and Fulton Mackay a fellow patient.
Like many a clown, Norman Wisdom could also play tragedy as those two sides are perhaps not so far apart. No coincidence that all three starring characters were/are known for their comedic roles.
Apart from the one issue of allowing you to save the space of a 2.5" drive, I cannot see what the point of this is. It would save far more space to omit the optical drive and stick to an HDD. Indeed you can exceed the capacity of any freely available optical disk format (50GB Blu-Ray, and I'm not sure RW media is available yet) with a 1.8" drive - they are available up to 320GB.
Optical media makes a pretty poor re-rewritable storage system for anything other than fairly large static files. If you want to transfer data, then generally a USB drive is going to be much more convenient. Just about the only obvious reason why you need an optical drive is for installing software, and in these days where a lot of software is simply downloaded, even that isn't critical apart from installing the OS in the first place (for which you can always use an external optical drive).
The comment about is inevitable is on the overall pattern, not any individual rider. Clearly there is a huge difference between different types of motorcyclists and I know a few who have been on two wheels and are not in their 50s. However, they tend to be the types who do a bit of touring or commuting as a cost effective and convenient way of travelling. They are still in more danger than somebody in a tin box (it's inherent in being more difficult to see and more vulnerable), but they are clearly not the ones at most risk.
That's very different in my experience to many of lovers of sports bikes. I'm not sure why anybody wants a 125bhp bike unless it's for thrills. There was a study in the US one different categories of bikes, and sports models were much more likely to be involved in fatal accidents than tourers or commuter bikes.
I passed my motorcycle test about the same time (1972 I think), on an even then, very old Lambretta LI150. Looking back at those years, before even compulsory crash helmets, I'm amazed that all my schoolmates survived. Onto University in London, then nobody could much afford personal transport apart from the twerp who smashed his Triumph Bonneville whilst demonstrating doing wheelies outside the hall of residence in South Kensington.
I too came across the later casualties, and can only imagine what regret some may have. it's especially tragic when this often happens in the young.
However, judging be the number of motorcyclists on the M4 on my commute to work who pass between lines of cars travelling at 70mph with inches to spare on each side, the tolerance (or even seeking out) of risk amongst many of the motorcycling fraternity is different to the aveerage person in their family hatchback. These sort of stories always bring out the Libertarians, but the statistics are stark. Per mile travelled, on average, a motorcyclist is over 30 times more likely to be killed than somebody in a car. Something like that adverse ratio is roughly reflected across the entire Western world.
More inherently more difficult to see, more unstable, much fast, much more vulnerable and often ridden by thrill seekers, the result is pretty well inevitable, with (in 2009) about one in 5 road deaths being of motorcyclists.
The unmarked cop car that pulled over the biker was not the one in the central reservation so how can you possibly know what the unmarked one that pulled the biker over saw? In any case, there could easily be other unmarked cars in radio contact.
The biker is a moron. Most likely he'll kill himself pulling some stunt or other. Pulling wheelies topping 100 mph on a public highway is madness.
Whilst I would certainly agree tha the current generation of storage array simply isn't up to getting the best out of the capabilities of an SSD as the internals and interconnects are simply not powerful enough, then I think there's good reason why storage arrays will not disappear and why a revolution in datacentre interconnect is going to happen any time soon.
The issue is around shared storage (rather than dedicated, which can be attached direct to server I/O buses). Sharing storage across multiple servers, whether it is by a shared array requires some form of network interconnect. It doesn't matter if that sharing is peer-to-peer, or central array. Something has to connect the storage services to the client systems. That network has to be capable of working over moderately long distances, perhaps 100metres or more, it has to be highly resilient, capable of dynamic reconfiguration without disrupting service, handle switching and routing of requests, and most importantly, must be widely supported - indeed at some levels, it has to be virtually ubiquitos. It also has to be very highly scalable.
The fact is that there aren't too many technologies to choose from for high performance shared network. You can forget wireless. It's far to slow and unpredictable. There are a few short distance interconnect technologies, such as FireWire, USB, SATA and so on that are moderately fast, but really not significantly more so than the main data centre interconnects which are, of course, Ethernet and Fibrechannel, Ficon and, or course, the converged Ethernet/FC standards (with a nod to Infiniband, although). There have been a few proprietary clustrer interconnect technologies, but these all have very limited support with nothing like open standards.
On those protocls above, we have real networks in the 10Gbps region available, with movement towards 100Gbps. In truth, there aren't many things out there which can properly support data rates like the latter. 100Gbps may only be the combined throughput of a dozen enterprise SSDs, but there's precious little out there that could deal with that amount of data so quickly as a single data source (aggregate bandwidth is another issue entirely).
I would argue that for most mainstream data centre applications, if the I/O latency dropped from the typical 6-8ms you might see for a random read on an array with entrerprise disks to the roughly 0.5ms you might expect to see with SSDs and a well designed modern I/O network and server technology, then that 100-fold improvement in latency will shift the bottleneck somewhere else. That's quite probably in processing, lock contentions or any number of other places.
Now there is no doubt that storage array suppliers are going to have to do something about the total throughput capability of their of the way they handle storage so that they aren't the bottleneck (even top end arrays can struggle once then get into IOP rates in the several 100s of thousands a second, or data rates in the 10Gigabyte per second region), However, I think for most data centre apps, an increase in capability of an order of magnitude would be ample, and we aren't in the territory where there would be much benefit in increases of 100 times that.
Of course there will always be the hyper-scale configs, like Google or (maybe) clouds, but those are also exercises in software engineering and very few organisationshave the need or resources for that.
I for one would be very suprised if, in 10 years time, the data centre storage interconnect didn't still rely on evolved versions of Ethernet and FibreChannel connecting to centralised storage facilities which may implemented a bit differently, but are still recognisable arrays. Datacentres evolve, they don't suddenly move to a new generation. The old has to work with the new. Arrays of some sort are here to stay.
I think the commenters on the original article were pretty well in agreement with the HP line on this (and one assumes those were not HP employees and, it appears, have practival experience of these issues).
That is that high capacity 3.5" FATA type drives have no real place in an array of this sort and that it is possible to achieve storage tiering using much more cost effective techniques than using up an expensive enterprise chassis with inherently low access density disks. There are "out of frame" expansion techniques that still provide the front-end functionality of the storage array. The 3.5" forem facor is dying as an enterprise format and if you look inside many of them you'll find the platter sizes are reduced in diameter in order to support higher spin speeds and reduced seek time. This is not a new thing - many of us will remember hard drives of 14" diameter. The need to reduce latency by higher spin speeds and shorter seek paths inevitably drives towards the smaller form factor.
The original article also allowed what looked like a pretty free voice to a competitor with a number of unspecific, but dismissive comments. I'm not suprised HP thought the original article wasn't balanced, and from what I can see, it misunderstood what enterprise arrays are generally used for.
"Though I am a little confused as to how two 70 kW power sources are supposed to supply four 145kW motors"
They clearly can't - at least in full. Clearly any such car on turbin alone will "only" get 140kW, but I suspect that will be enough to get the things to maybe 120-130mph on a flat road, which is way faster than the UK speed limit. Anybody thinking they can run at 150mph for hours on end on a cross-continental journey is going to be disappointed. However, just how many times is this even possible. Even on the German autobahn's there are lots of controlled sections, there is other traffic not to mention the sheer insanity of passing trucks and cars with a closing speed of 80Mph or not.,
No, the sensible way to view this is that the batteries will act as a reserviour for relatively short periods of acceleration. It is not a Le Mans racer.
Surely this concept is hugely preferable to a Tesla given the bad experiences that some of the testers have had on even quite modest trips with running out of battery charge.
Of course this thing is hugely heavier (probably), faster (at least on top speed) and is undoubtedly going to cost a lot more than the Tesla, but if it can be made to work, then you could even imagine using this on a long journey. The turbines no doubt cost a lot, but then so do the batteries in a Tesla.
There is also, of course, not comparison between the range under electric power and that of the fuel tank. The latter is simply a convenience issue for most people of how frequently you have to fill up (I don't see this thing crossing trackless deserts). The range of a purely electric car is a completely different issue as with contentional batteries you are in for a very long wait of many hourse before you can set off again, and that's assuming you can find a suitable high-power electric supply.
I recall a story from the late and much lamented John Peel. Finally giving up the arguments with his wife on direction finding on a holiday trip in Scotland, he decided on the quiet life and to just follow his wife's map reading instructions without question. He reported that they ended up on the coast, it just happened to be the one on the opposite side of the country to their intended destination.
No doubt I'll be in trouble with Ms. Bee for this annecdote reinforcing the outdated stereotype of hopeless feminine direction finding.
"it may be that the X3 simply flies mostly as an aeroplane at higher speeds, pitching its rotor blades flat to the air and letting them spin freely. "
Letting the rotor spin freely would also allow it to gain lift by acting as an autogyro with the appropriate ptich. I'm not sure those stubby wings will keep something that size aloft at 220knots. Nobody has flown an Autogyro at even twice the speed talked about here, but maybe it's possible by gradual changing the pitch and feathering.
Sensitive personal data has no place on any server exposed to the Internet in this way, and especially not on a web server, as it's only one remove from being hacked. Such information should be held on a further server hidden behind a firewall and only accessible via a secure network or VPN using strong authentication. Of course this law company is a very small operation which probably lacks the technical resources to implement such a thing, although the hosting company would (I hope) be able to put something appropriate together.
However, this is going to happen more and more as sensitive personal information becomes available to many smaller companies. The security is only as good as the weakest link.
Nice little irony here - try comparing SUN memory prices with the third party compatible (roughly 4:1 ration). Of course put the latter in and you are going to imperil your hardware support, not just for the memory, but for the whole machine.
Of course, SUN are not alone in this. It's standard practice for comparable manufacturers.
"gives priority to downhill routes"
That appears rather clever as I always though the net ascent or descent in a route is dictated solely by the difference in altitude between the start and end points. Whenever I've cycled between two points approximately as the same hight above sea level, I've always been rather of the impression that more downhill has also mean a compensating amount of uphill. As the latter takes rather longer to complete than the former, then the net pain appears to increase.
Personally I've always looked for the routes that involve the least amount of climbing and let the downhill parts sort themselves out. I will now have to search for these legendary downhill all the way routes, especially the circular ones, along with looking for those damned elusive magnetic monopoles.
Here's an idea. Why don't the manufacturers of routers put this functionality into one of their devices? OK - they may have to do something rather different so that they don't have a separate power brick, but it would save yet another box and cable if we could have one box that acted provided wired LAN, Wireless and PowerLine.
Having extensive experience of these things, the only way that you can scale and XP24K out to 2.26PB is to use 2TB FATA drives. Putting those in an enterprise array of this type is not, generally, a great idea. The reason for this is that they are first of all they support very low IOP ratings - half that of a 15K drive. This is compounded by being 4 times larger so the access density is 8 times worse. that the 500GB 15K drives. It gets worse - they HP documentation on these FATA drives state that they can only be used for random IO patterns 30% of the time, or you suffer abnormally high failure rates. Take that at face value, and you have only about 2% of the access density per GB on the FATA drives compared to enterprise for random IOPs (and actually managing the IO access pattern on a huge, shared enterprise array is a tricky thing to do).
If you want (relatively) low access rates on large amounts of data, then it's generally more cost effective to put those disks into an EVA8400 or the like. However, take that duty cycle seriously and be very careful what you use those FATA drives for. Bear in mind that it can take the best part of 24 hours to rebuild a RAID-5 set with 2TB drives (the only way you can max the usable storage) and, all that time, you are vulnerable to a second failure.
ftp://ftp.compaq.com/pub/products/storageworks/whitepapers/5982-7353EN.pdf
I do not know anybody who has maxed out and XP24K with FATA drives, and they would be mad to do so. If you use 600GB 15K drives in an XP24K then the total capacity is not much different to the P5000 and it occupies much less space and uses less power per GB for equivalent capacity and IOP count. If you need more than one chassis, buy a second one.
On any case, it is possible to use those EVAs and larger, slower disks as back-ends to boost the apparent capacity of the array.
Enterprise Arrays are not just about the absolute amount of disk capacity you can put in a box. In fact I would argue that isn't even in the top two or three priorities. First is data security and second is interruptible availability. Connect a few hundred servers to these things and you might as well pull the power supply in your data centre. Then there is maintaining high levels of reliable throughput and low latency on a mixed and changing workload so that when your data warehouse kicks in it doesn't screw up the response time in your call centres. Then you need those added value services - snapshots, array replication. Then you need capacity scalability, but you can (usually) always achieve that, with some compromises, by mapping across multiple arrays (the compromises are about ease of data migration and some array based async replication tools - and some manufacturer's arrays can cope with that). That dealt with, give me two smaller arrays with lower total power consumption and higher IOP count any time than the same capacity with some sub-enterprise disks.
Incidentally, as far as front-end FC interfaces go, there is not an Enterprise array manufactured which can approach the total bandwidth. That's a fact - all these large arrays run out of puff long before they can saturate all that FC capacity. The difficult bit is to balance the damned things so that you don't get hot spots in the SAN fabric, controllers or individual FCs.
So there is a lot more to what makes an enterprise array than the statistic of how many GB you can stuff into it, and a well-designed array will allow you to scale sideways.
Of course it is not possible for a ship to travel as fast as a train, but the fuel efficiency of the former per tonne-km (over long distances) is considerably better than that of moving freight by train. Look up the statistics and it is as much as twice as efficient for inter-continental type distances.
There are several reasons for that. Virtually any railway route covering thousands of km is going to have to traverse a mountain range. Hauling freight uphill costs energy. In contrast, oceans have the useful feature of being flat (or effectively so, with big ships an moderate seas). Then there is the very speed of rail which itself generates more friction and drag (and at relatively modest speeds - start pushing towards high speed train territory, and the fuel consumption goes up disproportionately. Then on rail there tends to be at degree of stop/start, and accelerating a few thousand tonnes to operational speed takes considerable energy (OK - there are regenerative braking systems on some electrified lines, but there is still a considerable loss of energy).
As a simple example, a horse can easily pull a 35 tonne narrow boat. Indeed a couple of men can do it, albeit at a modest speed. That would be utterly impossible on rail.
Then there is the embedded energy in the infrastructure and its maintenance. That's a huge amount on a permanent way of several thousand miles. In contrast, with shipping, you only have the docks at each end of the journey to worry about. The oceans in the middle are there for free.
If you want to move large amounts of bulk cargo, albeit relatively slowly, then ships are far an away the most energy efficient way of moving them.
Now I don't know how large the national DNA database is, but the amount of data held for any on profile is relatively small. All that is held is a limited number of markers - maybe just a few dozen per person. This means, in principle that you could put the raw DNA profile data from the approximately 3.1m people on the UK NDNAD onto a single 50GB Blu-Ray disk, although I've no doubt by the time that they've puffed this out with all the other overheads and information that is an operation system, it's much bigger than that.
Of course if what you mean is a full genome sequence for every individual, then that's a very different matter. With about 19,000 genes in the human genome, then any database size is going to be perhaps three orders of magnitude larger than one of mere DNA profiles.
Far and away the fastest way to move 60 million files is to use an image copy. Of course you can't restructure the underlying disk paritions, but it avoids all those tens of millions of file directory operations. You can then use a file sync program, like rsync to get the new and old back in full alignment (assuming that you aren't able to freeze the source file system in the meantime).
Depending on the total size, network bandwidth between the two servers and physical distance you can transfer the partition image using anything from a USB external drive to a network mount.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
