Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Undiscovered Country of HFT: FPGA JIT Ethernet Packet Assembly

timothy posted 1 year,4 days | from the fifo-with-a-vengeance dept.

Networking 452

michaelmalak writes "In a technique that reminds me of the just-in-time torpedo engineering of Star Trek VI: The Undiscovered Country, a company called Argon Design has "developed a high performance trading system" that puts an FPGA — and FPGA-based trading algorithms — right in the Ethernet switch. And it isn't just to cut down on switch/computer latency — they actually start assembling and sending out the start of an Ethernet packet simultaneously with receiving and decoding incoming price quotation Ethernet packets, and decide on the fly what to put in the outgoing buy/sell Ethernet packet. They call these techniques 'inline parsing' and 'pre-emption.'"

cancel ×

452 comments

Sorry! There are no comments related to the filter you selected.

Wow, (2, Insightful)

Antony T Curtis (89990) | 1 year,4 days | (#44961187)

This is the madness of high-speed trading...

Re:Wow, (0)

Anonymous Coward | 1 year,4 days | (#44961247)

This is the madness of high-speed trading...

We should be surprised? Porn has driven more technological advances than most will ever care to admit or acknowledge.

I'm certainly not shocked over extreme designs like this that drive billions of dollars per second. Even porn can't compete with that.

Re:Wow, (5, Insightful)

Anonymous Coward | 1 year,4 days | (#44961345)

At least porn is of arguable value. HFT brings absolutely nothing of value to the table. It doesn't help the traded companies, it doesn't help the market, it doesn't help any country economies. In fact, all it does is give the same hedge fund bozos who trashed the US and EU's economies another way to scarf income without adding anything.

Markets need to have some sanity. Either only allow trades each 15 seconds or tack a very small surcharge per transaction which wouldn't affect normal transactions, but penalize HFT enough to not make it worth the bother.

Re:Wow, (0)

Anonymous Coward | 1 year,4 days | (#44961423)

This analysis is correct.

Re: Wow, (4, Informative)

smaddox (928261) | 1 year,4 days | (#44961773)

Why only 15 seconds? Why not 24 hours? What reason, other than gaming the system, could there ever be to hold a stock for less than 24 hours? I don't understand why this wasn't made illegal 30 years ago... Well, I do - the people making the laws are the people profiting from them - but the reason is not a good one.

Re:Wow, (5, Insightful)

vux984 (928602) | 1 year,4 days | (#44961967)

I agree with you, but go further.

HFT brings absolutely nothing of value to the table. It doesn't help the traded companies, it doesn't help the market, it doesn't help any country economies. In fact, all it does is give the same hedge fund bozos who trashed the US and EU's economies another way to scarf income without adding anything.

HFT isn't merely neutral adding no value; the income they "scarf" is income the rest of the stake holders lose. (To the extent that one can lose something one never got, at least.)

Either only allow trades each 15 seconds

I propose 10 minutes or even longer, and even that's more than fast enough. A company's fundamental value doesn't change 6 times an hour.

10 minutes lets news hit, lets people think and consider what they value the stock at, and even people not day trading for a living can react and put in a trade order without being 100 million trades "too late".

or tack a very small surcharge per transaction which wouldn't affect normal transactions, but penalize HFT enough to not make it worth the bother.

Not just every transaction -- every ORDER. HFT spams millions of orders to probe, guide, bait, etc, most of them never close and are cancelled; within milliseconds of being placed.

Those need to be 'taxed' as well.

Finally eliminate "dark pools". All trades MUST go through regulated markets. There should be no dark pools of unregulated trade.

Re:Wow, (0)

Anonymous Coward | 1 year,4 days | (#44961315)

It sounds a lot like MacOS - pre-emption and inline-parsing, we've had that for years.

Re:Wow, (1)

dintech (998802) | 1 year,4 days | (#44961461)

However, no-one ever wrote a trading platform on MacOS.

Re:Wow, (-1)

Anonymous Coward | 1 year,4 days | (#44961883)

dude, seriously? MacOS ~= FreeBSD ~= UnixWest....UnixEast => Linux. 100% of HFT systems run on Linux. Just like 100% of the 500 fastest machines on the planet run on neither Windows nor MacOS. It's not the OS, sweetheart.....

Re:Wow, (4, Insightful)

Anonymous Coward | 1 year,4 days | (#44961327)

High speed trading is all about stealing as much money as quickly as possible before anybody else has a chance to do it.

It serves no purpose but to move money from the hands of everyone else to the banks -- the same banks that caused the financial mess in the first place.

I blame this squarely on the American concept of capitalism -- the Republicans and Libertarians will tell you this is how it's supposed to work, because they and their benefactors directly profit from this.

The super rich just skim off the top and bypass the entire 'market' for their own purposes.

And this is why there never has been, and never will be a free market and it can't do what it claims -- because someone will always game the system for their own benefit, and there's nothing to keep them in check. And the big players define the market choices, so there is no actual options for people.

I view HFT as wholesale theft, which is pretty much exactly what it is. Nobody earned anything, they just have a direct hook into the system whereby they can bypass everything and make money before it exists to anybody but them.

Burn Wall Street. Eat the rich.

Re:Wow, (2, Informative)

Anonymous Coward | 1 year,4 days | (#44961459)

That's right!
I understand that none of the Democrats will have anything to do with WallStreet.
Heck, not a single one of them owns any stocks from what I hear.

Re:Wow, (1)

gmuslera (3436) | 1 year,4 days | (#44961729)

The madness goes above it. In a system where the ultimate good is money, going all for it is a reasonable move. Making dedicate machines that make money as fast as possible is a good approach. The madness there is that the ultimate good is money (for everyone, even the ones that make the laws), more than human life or that mankind or civilization have a future.

What a waste (2, Insightful)

stewsters (1406737) | 1 year,4 days | (#44961201)

These people could be used to develop more efficient hardware for everyone to use, or fix medical conditions, rather than make rich traders even richer at the expense of another economic collapse. It seems wrong that our economy prioritizes high frequency trading so much.

Re:What a waste (5, Funny)

larry bagina (561269) | 1 year,4 days | (#44961309)

You could be used to develop more efficient hardware for everyone to use, or fix medical conditions, rather than posting on slashdot.

Re:What a waste (1)

i kan reed (749298) | 1 year,4 days | (#44961385)

It seems wrong, because it reflects a fundamental lack of concern for what securities represent. It is another step on the road of investment being more valuable than insight, labor, or creating meaningful value.

The underlying purpose of the stock market is the valuation of companies and what they do, but we see more and more evidence that the metadata outweighing the data. It's a subtle incrimination of corporate capitalism, that those involved really don't want to acknowledge.

Re:What a waste (1)

Ralph Wiggam (22354) | 1 year,4 days | (#44961493)

You can have a room full of insightful, creative people. But without investment, they won't be accomplishing much.

Re:What a waste (1)

sjames (1099) | 1 year,4 days | (#44961631)

OTOH, HST is not investment at all. Th market has become so messed up that it's easy to get investment for mail order pet food, but not for useful technology.

Re:What a waste (1)

the eric conspiracy (20178) | 1 year,4 days | (#44961757)

That has nothing to do with the market and everything to do with the fact that the US is a consumer driven society.

Selling pet food brings an immediate, strong and much more certain return on capital than developing a new technology.

Re:What a waste (1)

sjames (1099) | 1 year,4 days | (#44961917)

Except it didn't. It was just another malinvestment.

Re:What a waste (0)

Anonymous Coward | 1 year,4 days | (#44961947)

you are so full of shit.

Re:What a waste (1)

the eric conspiracy (20178) | 1 year,4 days | (#44961679)

Effective allocation of capital is absolutely critical to an efficient economy. It is a sine qua non of free societies.

It is why centrally planned economies don't work in competition with market economies. Look what happened to China when they released some the the reins vs. Mao's 5 year plans.

Why do you think the credit collapse had such a dire effect on the world economy 5 years ago?

Labor and creativity are important too, but they don't turn into meaningful economic activity without capital.

Investment is fundamentally important and always will be.

Re:What a waste (1)

Alioth (221270) | 1 year,4 days | (#44961407)

Maybe so or maybe not. A bit like the arguments "We shouldn't have both Gnome *and* KDE! Everyone should work on just one project so we get more done". The thing is maybe people working on HFT systems don't want to work on medical devices and wouldn't work on medical devices even if HFT didn't exist. I don't happen to work on HFT systems myself, but personally I wouldn't want to work with the stifling regulations there are for medical devices, I'd rather work on something that's rather more free in how you develop things.

Once you start saying things like this, you can also say "Well, no one should write video games. We shouldn't be wasting our time having people write Starcraft 2, they should all be working on medical devices".

Re:What a waste (1)

h4rr4r (612664) | 1 year,4 days | (#44961439)

You think they do it for the love of HFT?

For that kind of money I would gladly deal with medical device regulators all day.

Re:What a waste (1)

internerdj (1319281) | 1 year,4 days | (#44961559)

There are some pretty interesting technical challenges with HFT and a competitive problem that isn't inherent to most technical problems which tend to often be man versus nature. It could have some appeal. That said, there isn't an amount of money and set of technical challenges interesting enough to make me want to work for that type of management structure and corporate environment.

Re:What a waste (3, Interesting)

Anonymous Coward | 1 year,4 days | (#44961433)

They have. See https://en.wikipedia.org/wiki/Alphamosaic for their previous jobs. Also, 3rd generation of their chips is in the Raspberry Pi, as the CPU.

Re:What a waste (-1, Flamebait)

Minwee (522556) | 1 year,4 days | (#44961511)

Stop that. When you start bringing "facts" into a Slashdot discussion you get in the way of the raw, unbridled slacktivist faux-outrage that makes life worth living.

You know who else talked about what Alphamosaic did? Hitler.

Re:What a waste (0)

Anonymous Coward | 1 year,4 days | (#44961555)

I couldn't have put it better myself. Talking about a complete waste of time, money, and talent. Just sad.

Re:What a waste (0)

Anonymous Coward | 1 year,4 days | (#44961735)

There's no evidence HFT contributes negatively to the market, let alone that it will be the catalyst of an impending economic collapse. Those that feel otherwise have fallen prey to the latest leftist reactive propaganda to an emerging and innovative free market that they fail to grasp. On the other hand, there's evidence that HFT provides added liquidity to the market and lowered costs, which is beneficial. Not to mention, as with any free market, if the participants didn't benefit, they would cease to participate. Therefore, your assumption that HFT is a wasteful use of talent and resources is proven invalid. Further, in a free society consenting adults are free to buy and sell from each other as they wish, despite the tyrannical views of some.

Re:What a waste (1)

gandhi_2 (1108023) | 1 year,4 days | (#44961781)

Unlike the noble persuit of high-end graphics cards to build a totally bitchin' mining rig.

Re:What a waste (1)

Bengie (1121981) | 1 year,4 days | (#44961871)

I see it like racing. Lots of money goes into making fast cars go faster for something that has no real value. Yet many of the technologies that go into race cars still make their way into regular cars.

These kind of situations making crazy ideas worth trying and lots of money to back it. How long until home users get the benefit for FPGA assisted networking because it becomes a well-know subject and gets crazy cheap to implement because some rich people wanted to try to make even more money in trading?

Re:What a waste (1)

nurb432 (527695) | 1 year,4 days | (#44961927)

Without wealthy people you wont get your research for 'everyone else to use'.

Next step : superluminic packet (3, Interesting)

aepervius (535155) | 1 year,4 days | (#44961203)

So that you can send your order before even the price quotation comes. Oh , wait, those guy already can do that and send packet back in time : http://news.slashdot.org/story/13/09/25/1955220/somebody-stole-7-milliseconds-from-the-federal-reserve [slashdot.org] /sarcasm

Re:Next step : superluminic packet (1)

necro81 (917438) | 1 year,4 days | (#44961357)

There has been speculation of people developing neutrino-based communication systems for high-speed trading. Because neutrinos pass through matter more-or-less unabsorbed and unimpeded, and travel at more-or-less the speed of light, you can do line-of-sight communication straight through the Earth. So, for instance, a trader in NY can send buy/sell orders to Japan by beaming neutrinos through the Earth and have them arrive faster than orders sent via optical fiber over the surface of the Earth (which is a longer path, and the speed of light in optical fiber is less than c in a vacuum), or via satellite.

Re:Next step : superluminic packet (1)

ebno-10db (1459097) | 1 year,4 days | (#44961409)

Neutrinos? Bah. A wormhole lets you shrink the distance between 2 points, so the speed of light is less of a problem.

Re:Next step : superluminic packet (4, Informative)

Anonymous Coward | 1 year,4 days | (#44961523)

Neutrinos exist.

Re:Next step : superluminic packet (1)

pla (258480) | 1 year,4 days | (#44961649)

AC or not, this post count as 100% the correct response to the GP. Why did it get modded to -1?

Re:Next step : superluminic packet (1)

Minwee (522556) | 1 year,4 days | (#44961563)

So, for instance, a trader in NY can send buy/sell orders to Japan by beaming neutrinos through the Earth and have them arrive faster than orders sent via optical fiber over the surface of the Earth (which is a longer path, and the speed of light in optical fiber is less than c in a vacuum), or via satellite.

But even if they were using tachyons for communication, their orders would never arrive before the ones which come from a rack that is only five metres away from the exchange and connected directly to it with a straight optical cable. Even if the trader is sitting in NYC, they can still run their algos in Tokyo and get the jump on any other traders who don't.

Why resort to science fiction when there is already a better way?

Re:Next step : superluminic packet (0)

Anonymous Coward | 1 year,4 days | (#44961725)

Because sometimes the prices are different in Tokyo and NYC. If there's a price rise in NYC, and that signal travels via fiber-optic cable to Tokyo, then there's a few milliseconds when you could buy in Tokyo and sell in NYC and make an instant profit. That requires you to have communications faster than the exchange and faster than anyone else.

So much innovation for so little value (3, Insightful)

hirschma (187820) | 1 year,4 days | (#44961205)

Things like high frequency trading make me want to vom. Essentially, all they're doing is shuffling money around, taking advantage of an outdated system, and increasing risk for the entire world.

It'd be great to see this kind of innovation in something that actually is useful and valuable - not for creating an incremental improvement on a corrupt system.

Re:So much innovation for so little value (4, Insightful)

KiloByte (825081) | 1 year,4 days | (#44961249)

Essentially, all they're doing is shuffling money around.

More exactly: they're shuffling money from our accounts to theirs. If your bank has any securities involved with the stock exchange, it is your money that is getting nicked here. A zero-sum game requires that besides a winner there must be a loser.

Re:So much innovation for so little value (5, Insightful)

HornWumpus (783565) | 1 year,4 days | (#44961341)

The big loser is the trading clearinghouse/broker. Bid/ask spreads are about 1% of what they used to be.

The other big losers are speculators (as opposed to investors, look up the difference).

I see no problem with this except when the HFtraders are able to say: 'Our bad, back out all those losing trades for us'. That's just fucked.

Re:So much innovation for so little value (1)

i kan reed (749298) | 1 year,4 days | (#44961425)

Except HFT is causing bigger spikes on stock prices, helping the speculators and hurting the investors who decide to get in or out at the wrong time.

Re:So much innovation for so little value (1)

the eric conspiracy (20178) | 1 year,4 days | (#44961567)

There is no evidence of that. All the really big stock price events occurred in the past. Like the 22% one day drop in 1987. In fact it's likely HFT reduces large spikes because the arbitrage occurs much more quickly.

And also just so you know, investors don't try to time markets. They hold for long periods of time so these fluctuations don't matter.

Re:So much innovation for so little value (1)

Bengie (1121981) | 1 year,4 days | (#44961893)

If you an investor buying during a spike, you're doing it wrong.

Re:So much innovation for so little value (2)

AmiMoJo (196126) | 1 year,4 days | (#44961471)

Just wait until someone figures out how to write an exploit targeting HF trading systems. Within milliseconds they could execute the biggest cyber attack in history and crash the economy of an entire country. They could even interlock it with trades from other countries just to make un-picking all the damage nearly impossible.

Of course it might just happen that way due to a genuine bug in one of the systems. Would be hard to tell which it was from a distance.

Re:So much innovation for so little value (1)

Ralph Wiggam (22354) | 1 year,4 days | (#44961569)

Your basic knowledge of trading has no place here.

Keep in mind that you just replied to someone who thinks copyright laws are a "crime against humanity".

how long before this house of cards collapses? (0)

Anonymous Coward | 1 year,4 days | (#44961355)

Ya know, as a Morlock, I'm starting to look at them less as gated communities and more as veal pens.

Re:how long before this house of cards collapses? (1)

ebno-10db (1459097) | 1 year,4 days | (#44961419)

Ya know, as a Morlock, I'm starting to look at them less as gated communities and more as veal pens.

Veal? That's barbaric. Wait until they're older - they're still pretty tender at 18.

Re:how long before this house of cards collapses? (0)

Anonymous Coward | 1 year,4 days | (#44961565)

fine. "feed lot" then. For some reason, "veal pen" popped into my mind first. More ghoulish, see, poster really is a Morlock.

Re:So much innovation for so little value (0)

Anonymous Coward | 1 year,4 days | (#44961623)

Um, yes it does -- but the equities market isn't a zero-sum game. Futures, options, and many other asset classes are, but not equities.

Also, "from [your] accounts to theirs"? Only if you wanted to trade and got filled at a worse price as a result of HFT activity. The HFT guys would argue that they increase liquidity and reduce bid/ask spreads, which may well have gotten you a better price than you could have had in the pre-HFT days. Not sure I agree, and there are other good reasons to criticize HFT, but your arrows are off the mark.

Re:So much innovation for so little value (3, Insightful)

green is the enemy (3021751) | 1 year,4 days | (#44961657)

I think most non-finance people who read a little about high-frequency trading systems are repulsed by the concept. Can someone point out a useful economic function for them?

One the one hand they dramatically reduce the bid-ask spread for low-volume trades. But is that really all that economically beneficial?

On the other hand, they prey on large trades that mutual funds must place to rebalance their holdings. This probably hurts the average investor quite a lot.

Some claim that they increase liquidity and reduce volatility. Is that really so?

Re:So much innovation for so little value (0)

Anonymous Coward | 1 year,4 days | (#44961739)

There's a less well known but disturbing corollary to "when someone wins someone must lose"

When someone loses, someone must win.

The rich have figured out that it's much easier simply to sabotage the economic well being of the lower/middle class than earn money the hard way. Every time you hear about politicians axing services, social safety nets, removing employee protections, and crying about taxes.. This is what they are doing. They are attacking you. Someday I hope the public decides to fight back.

Re:So much innovation for so little value (0)

Anonymous Coward | 1 year,4 days | (#44961801)

Active vs. passive investing (see Bogleheads). Passive investors don't lose money to HFT. Only active investors who are trading against HFTs are losing money to them.

Want to not get screwed by HFTs? Stop trading frequently yourself, and hold your securities longer term.

Re:So much innovation for so little value (0)

Anonymous Coward | 1 year,4 days | (#44961475)

Things like high frequency trading make me want to vom [..] It'd be great to see this kind of innovation in something that actually is useful and valuable.

I suspect that if you're the kind of person who works on technology designed to aid HFT, your motiviation in life is quite plainly money- or (giving such people more benefit of the doubt than they deserve) the chance to work on interesting tech without giving a toss what it's actually used for.

It's clearly *not* doing anything that adds value to the human race- quite the opposite. Am I the only person who thinks that if all those working on HFT fell under a bus tomorrow, the net benefit to humanity would be positive?

I don't give a toss if you're a genius with fantastic tech skills- if you're applying them to a system that only benefits the entrenched interests of those exploiting their size and dominant position (in a mockery of the free market), against everyone else, you're worthless. Throw yourself under a bus- there's a number 20 due in 5 minutes time.

Re:So much innovation for so little value (1)

gmuslera (3436) | 1 year,4 days | (#44961833)

You are playing their game too, you just get outraged because they play it more efficiently. They played enough with the rules to make millons just moving aluminum between 2 warehouses [nytimes.com] . Life is more than money, but for our current culture is everything.

Re:So much innovation for so little value (0)

Anonymous Coward | 1 year,4 days | (#44961969)

Essentially, all they're doing is shuffling money around, taking advantage of an outdated system, and increasing risk for the entire world.

What could go wrong with this? I'm sure they've tested every possible scenario and have lots of fail-safe mechanisms built-in (at least mechanism that protect their assess... I mean assets.

Huh? (0)

Anonymous Coward | 1 year,4 days | (#44961219)

This is investing?

Re: Huh? (4, Insightful)

Anonymous Coward | 1 year,4 days | (#44961301)

No, this is arbitrage. Taking advantage of price differences within or between markets.

Re: Huh? (0, Troll)

sxpert (139117) | 1 year,4 days | (#44961331)

technically, it is theft.

Re: Huh? (1)

gl4ss (559668) | 1 year,4 days | (#44961389)

technically it's more like bribing the telegram machine operator to perform trades based on rules you give him..

why bribing? well you have to know the right people to pay money to even begin trading in this fashion. now I wouldn't have a problem with it if it was accessible to everyone, for example if anyone could buy machine time from vm's that were all given the information at the same time(artificially arranged, wouldn't work otherwise!) at the stock exchange.

Re: Huh? (3, Insightful)

Minwee (522556) | 1 year,4 days | (#44961765)

now I wouldn't have a problem with it if it was accessible to everyone, for example if anyone could buy machine time from vm's that were all given the information at the same time(artificially arranged, wouldn't work otherwise!) at the stock exchange.

Sure you would. You might not think so, but suppose that the exchange set up a perfectly equitable system in which identically configured were made available to every firm and provided with identical market feeds all perfectly synchronized so that no single trading VM has any advantage over any other.

I would give that system about three hours of run time before you discover that:

  • - Your competitor has just acquired seventeen different vms through deals with other firms and is using them to dominate the market, pushing out firms using only one or two vms,
  • - Your market feed is being saturated with bogus bids and offers which reduce your ability to see real market data, but somehow one of your competitors is able to filter it all out and trade ahead of you,
  • - Other VMs on the same host as yours suddenly start running at 100% of their CPU and I/O capacity, causing slowdowns for you, and
  • - A mysterious and completely untraceable hardware fault causes all vms except for the first seven to experience periods of unexplained latency at key times during the day.

Just look at the tricks that players in the game are already using [nanex.net] , and ask yourself how changing the rules is going to stop them.

Re: Huh? (0)

Anonymous Coward | 1 year,4 days | (#44961399)

To-may-to, to-mah-to...

Re: Huh? (0)

Anonymous Coward | 1 year,4 days | (#44961707)

Um, no.

Lets say someone tells you that they are willing to pay $10 for an item. ( a bid)
You check all the available markets, and notice that some particular vendor is offering such an item for $9. (offer)
You purchase the item for $9, sell it to your friend for $10 and pocket the $1. That's arbitrage.
You will notice that everyone was paid. Nothing was stolen.

Since you did the effort to find the deal, you get to keep the difference. Your friend is happy to pay you the extra dollar because he was too lazy to look himself and still got what he wanted for the price he wanted. The person selling it to you for $9 is happy because they were wiling to sell at that price.

There are, certainly, lots of problems with the financial market. But calling all arbitrage "theft" is absurd..

Fuck off and die (2, Insightful)

Anonymous Coward | 1 year,4 days | (#44961223)

Fucking parasites (and their toolmakers).

Your Start Trek reference (2)

Anonymous Coward | 1 year,4 days | (#44961225)

Makes no sense.

Re:Your Start Trek reference (0)

Anonymous Coward | 1 year,4 days | (#44961295)

Your comment subject makes no sense.

Re:Your Start Trek reference (1)

BravoZuluM (232200) | 1 year,4 days | (#44961405)

Sure it does. Just like the torpedo, this board will single-handedly cause a market explosion by with all of the other trading platforms can target. It will be glorious.

Older idea than you might think (4, Interesting)

Anonymous Coward | 1 year,4 days | (#44961279)

I have a friend who works in the defense industry but interviewed with HFT firms around 3 years ago. My friend also had this idea, and discussed it with me since I am close to the industry.

If an industry outsider like my friend had this idea within a week or two of merely interviewing for jobs, it is a good bet many others had already conceived it and even gotten it working before that.

Re:Older idea than you might think (1, Insightful)

Anonymous Coward | 1 year,4 days | (#44961815)

I have a friend who works in the defense industry but interviewed with HFT firms around 3 years ago. My friend also had this idea, and discussed it with me since I am close to the industry.

If an industry outsider like my friend had this idea within a week or two of merely interviewing for jobs, it is a good bet many others had already conceived it and even gotten it working before that.

Yes, and when I was 8 I had the idea for a nuclear-powered photon drive for a rocket ship. Unfortunately translating the concept from my crayon drawing of a Saturn 5 with light-bulbes instead of exhaust nozzles to a working prototype proved to be impractical.

Ideas are cheap. Implementations are what matters. Particularly for something like this where hardware and software both need to be developed and speed and reliability are paramount. Having a basic idea of the algorithm is a far cry from a devise that works reliably enough and faster enough to be an improvement over established competing systems.

Take a penny... Take a penny (0)

Anonymous Coward | 1 year,4 days | (#44961283)

Someone's gotta leave a penny...

Solution for a non-problem (1)

depressedrobot (1067078) | 1 year,4 days | (#44961329)

Ok I can see it having implications in high speed control applications now that I think of it. Something like CERN, NIF or any space agency could use it. But we wont have any money to build anything because its more profitable to build shit like this to skim parts of pennies out of the jar than to actually build cool shit that needs FPGA controlled switches.

Jerk warefare (0)

Anonymous Coward | 1 year,4 days | (#44961339)

These jerks doing HFT is giving people in the financial sector a bad name. An ethical and moral person wouldn't do this type of shit, since it's clearly wrong. Yes, it is currently legal in that the jerks have paid off the government, bought the laws they want and stripped the SEC of their power.

No matter what Wall St apologists say, this is skimming. There's no polite way to put it. The argument that HFT creates liquidity is a complete lie. All it does is shuffle money from 90% of the people trading into the pocket of 1%.

Re:Jerk warefare (1)

ebno-10db (1459097) | 1 year,4 days | (#44961527)

These jerks doing HFT is giving people in the financial sector a bad name.

I didn't know they could get a worse name.

No matter what Wall St apologists say, this is skimming.

Probably so, but it's nothing compared to the outright criminal activities (especially control fraud) that were practiced pre-crash, and probably post-crash as well. The DoJ, SEC, OCC, etc., etc., etc. have bent over backwards to not investigate, let alone prosecute these crimes. For good measure, toss in the corrupt but unfortunately not illegal practices of the Fed engineering yield curves to benefit the banks, buying commercial trash securities (the law says the Fed is only allowed to buy high quality securities) or letting investment firms like Goldman-Sachs have access to Fed loan facilities (they're supposed to be limited to depository institutions).

Out, Out, Damned Glitch (2)

b4upoo (166390) | 1 year,4 days | (#44961351)

Just think of the expenses that could be involved if one of these programs screws up. Instead of sell! sell! sell! One could buy! buy! buy! the wrong stock in very large quantities. Instead of a billion dollars in buggy whips one might end up with a billion dollars in donkey whips. What a bummer.

Re:Out, Out, Damned Glitch (2)

Minwee (522556) | 1 year,4 days | (#44961847)

Just think of the expenses that could be involved if one of these programs screws up.

This has already happened [nanex.net] . The screw-ups in question lost over $400,000,000 in half an hour.

Re:Out, Out, Damned Glitch (1)

gmuslera (3436) | 1 year,4 days | (#44961861)

Is a no risk operation. If they screw up big, they will be bailed out again. The more important asset in their pocket is still the government, all versions of it.

HFT can easily be stopped (1)

Anonymous Coward | 1 year,4 days | (#44961361)

Set a minimum x hours before a stock can be resold.

Re:HFT can easily be stopped (0)

Anonymous Coward | 1 year,4 days | (#44961551)

Some markets only clear your account after 3 days past the operation, to ensure HFT and day-trading are kept to a minimum.

Not sure why this is still a big deal in the US.

Re:HFT can easily be stopped (0)

Anonymous Coward | 1 year,4 days | (#44961581)

Set a minimum x hours before a stock can be resold.

How many $$$ do you have to push that law?

It will have to be more than the financial industry can put on (or below) the table.

Re:HFT can easily be stopped (1)

boristdog (133725) | 1 year,4 days | (#44961593)

Set a minimum x hours before a stock can be resold.

You don't even need x hours, you could just make a rule that stocks need to be held for a few seconds, or put a random 1 to 5 second delay on each trade.

Re:HFT can easily be stopped (0)

Anonymous Coward | 1 year,4 days | (#44961785)

The best solution I've heard is to execute trades on something like a 1 second timescale. Basically, gather all trades, and match / execute them every 1 second. This would eliminate the sub-second delay advantages, and not add weird regulations to the market.

I'm not very experienced in this area, so I've no idea if it would actually work / help, but it is at least an interesting idea.

Re:HFT can easily be stopped (2)

Minwee (522556) | 1 year,4 days | (#44961953)

Okay, I am holding 500 shares of ABCD. At 10:01:32.0512 I buy 500 more shares at 10.12. At 10.01:32.7691 I sell the original 500 shares at 10.13, but am still holding on to the new 500 I just bought. Have I just broken your simple rule?

I have? Then how about this: Firm A buys, and then firm B sells the same thing less than a second later. X hours after the original buy, Firm A sells the same stocks to firm B. While both firms are owned and run by different people, they share the same address and make a large number of trades with one another. Are either of these firms breaking your rule which easily stops HFT?

Concepts are easy. Making rules is hard, especially for a game which is played by the biggest rules-lawyering munchkins on the planet.

Don't understand (2)

XanC (644172) | 1 year,4 days | (#44961485)

What is with the vitriol here? Why should buyers and sellers not be able to come together to make a transaction at any time that they like?

Re:Don't understand (1)

SirGarlon (845873) | 1 year,4 days | (#44961643)

The point of high-frequency trading [wikipedia.org] is for the buyer to pull a fast one: to offer a price that an informed seller would not accept, before the seller can find out what the fair price should be.

Re:Don't understand (-1)

Anonymous Coward | 1 year,4 days | (#44961983)

I am so glad you are spouting this extremely uninformed dribble.
It is better if you don't trade on the stock market at all.

Re:Don't understand (1)

pla (258480) | 1 year,4 days | (#44961789)

What is with the vitriol here? Why should buyers and sellers not be able to come together to make a transaction at any time that they like?

They should!

But computers can't "like" ice cream, either.

Re:Don't understand (1)

sjames (1099) | 1 year,4 days | (#44961809)

Because this is rent seeking and has a lot more to do with jumping in between legitimate buyers and sellers to extract money than it does with investment.

Also because the people doing it have already managed to crash the world economy once and through political maneuvering got a handsome payout for doing it rather than the jail time they deserved. Because in spite of the big handouts from the taxpayers they continue to shit all over everyone.

They should be roasted on a spit until they stop screaming.

This is nothing! (1)

Medievalist (16032) | 1 year,4 days | (#44961487)

In Chicago, traders are working faster than light [motherjones.com] .

You have to be a rich bankster to achieve faster than light trading, though. If anybody else does it it's cheating.

They cut off the end of the summary (1)

slashmydots (2189826) | 1 year,4 days | (#44961543)

Actual full summary ending: "They call these techniques 'inline parsing' and 'pre-emption' and 'greedy unfair asshole companies making billions for their owners through cheap, cheating tactics like this to undercut smaller startups and push them out of the market so their fat owners can buy another personal jet instead of handing off the profits to their investors/customers'"

Most tenuous link ever? (2)

wonkey_monkey (2592601) | 1 year,4 days | (#44961585)

In The Undiscovered Country they modified a torpedo to home in on gas emissions from a Klingon Bird of Prey. This is a story about building trading algorithms into an ethernet switch.

Apart from "needing to do something quickly," I really, really can't see the connection.

Re:Most tenuous link ever? (1)

XanC (644172) | 1 year,4 days | (#44961661)

The thing's gotta have a tailpipe.

Re:Most tenuous link ever? (0)

Anonymous Coward | 1 year,4 days | (#44961859)

IIRC, they were working on modifying the torpedo as it was being loaded into the launcher.

And this is a good thing because..... ? (1)

gsslay (807818) | 1 year,4 days | (#44961611)

Making lots of money by pushing network packets around faster, to no real net benefit to anyone. Other than the ones pushing the packets.

What would be the problem with instigating... (0)

Anonymous Coward | 1 year,4 days | (#44961633)

a minimum processing interval for trades, such that trading would only take place every X number of seconds/minutes/hours so as to eliminate the need for dedicated and rapid uplinks to the stock market in order to be competitive?

Other than perhaps eliminating HFT, I can't see a whole lot of major changes to the market and with longer intervals auditing trades would become much easier to do since there would be less data to sift through for any given period of time.

But maybe that's a simpleton's view of the situation.

Pipelining (1)

K. S. Kyosuke (729550) | 1 year,4 days | (#44961697)

They call these techniques 'inline parsing' and 'pre-emption.'"

And everyone else has been calling this "pipelining" for decades.

Re:Pipelining (1)

michaelmalak (91262) | 1 year,4 days | (#44961915)

And everyone else has been calling this "pipelining" for decades.

I'm not aware of any pipelining system that performs partial parsing. In CPU pipelining, "instruction decode" is AFAIK a single, atomic step.

Great.. (0)

Anonymous Coward | 1 year,4 days | (#44961827)

An interesting technique/technology, created for an utterly worthless and counterproductive "industry". I know they say that HFT (High Frequency Trading) is supposed to "increase liquidity" but it produces nothing and removes billions from the industries that actually do something (I would even include entertainment & service(s) industries in that). If I had to come up with an analogy to match my perception of HFT it would be an individual who spends tens of thousands of dollars to develop a device that they can bury at a stop light and siphon a little gas from each vehicle that stops over it. It provides nothing to the roadways (save for a little bit of new pavement), doesn't create any really useful technology (at least vs its cost) and basically steals from everyone who is unfortunate enough to stop over it that are likely going to work, transporting goods, or going to buy goods.

Who Stole 7 Milliseconds From the Federal Reserve? (2)

transporter_ii (986545) | 1 year,4 days | (#44961867)

I think we just found our culprit.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?