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!

Contemplating Financial Trading At Picosecond Resolution

timothy posted more than 3 years ago | from the race-to-the-bottom-yowza dept.

Businesses 448

pbahra writes "One complaint made of the modern stock market is that it is concerned too much on the short term. A second is a long time in cash-equities trading. Four or five years ago, trading firms started to talk of trading speeds in terms of milliseconds. But in recent weeks trading geeks have started to talk about picoseconds, in what is a truly mind-boggling concept: a picosecond is one trillionth of a second. Put another way, a picosecond is to one second what one second is to 31,700 years."

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

Worthless (4, Insightful)

Anonymous Coward | more than 3 years ago | (#35376512)

These people are parasites. They provide nothing of value to the world; they just take. This crap should be illegal.

Re:Worthless (0)

FooAtWFU (699187) | more than 3 years ago | (#35376596)

The thing about short-term trading is that it's really short term. It doesn't usually have much effect in the long term, and only matters a lot if you're trying to compete with them in day-trading. (The one notable exception, the "flash crash", was reasonably extraordinary).

And if they were really introducing inefficiencies into the market, the next picosecond trader in line would be all to happy to arbitrage the difference, one way or another.

Re:Worthless (5, Insightful)

Kenja (541830) | more than 3 years ago | (#35376658)

The money they take out of the system is real, it has an effect and their actions have consequences. What's more, the changes in the market that they are betting on and influencing have nothing to do with the real world. It amounts to influenceable gambling where the act of betting effects the outcome of the bet.

Re:Worthless (1)

johanatan (1159309) | more than 3 years ago | (#35376926)

They do provide liquidity and what essentially amounts to inertia (or a process of stabilization--excluding cascading failures of course). It's not hard to imagine that the market would fluctuate more in the absence of these micro-arbitrations.

Also, in the case of derivatives, there's always someone looking to buy or sell almost anything for many [legitimate] reasons [given their extant position] (and these traders make the long-tail market for such people).

Re:Worthless (5, Insightful)

copponex (13876) | more than 3 years ago | (#35376676)

Goldman Sachs has colocated at the NYSE, and is front running the stock market to the tune of 13.4 billion dollars in profit every year, simply because of their location. And they also sell self destructing financial instruments to their own clients while betting against them. Here, it's been in the news. [pbs.org] But I doubt you watch the news.

So, they're fucking cheating shits who do nothing but game algorithms and lie to people to steal their money, and you're a stupid cunt for having such blind faith in an opaque market.

Re:Worthless (0, Interesting)

Anonymous Coward | more than 3 years ago | (#35376862)

They're silently inserting themselves as a middle man, making your trades just a little bit worse off for you. That behavior is incompatible with a free market. I believe there's a saying: no single snowflake is responsible for the avalanche. Or, in terms you could understand, no single sperm is responsible for the stinging in your eyes after a half dozen niggers make you the center of a bukkake session. And yet, we're still feeling the effects of a financial crisis caused by individual actions which don't usually have much effect in the the long term.

Re:Worthless (3, Interesting)

Myopic (18616) | more than 3 years ago | (#35376698)

I don't think it should be illegal, I just think one dimension of the assessed taxes should be length of time the asset is held. Set thresholds at, say

1 second -- 99.99% marginal tax
1 minute -- 99% marginal tax
1 hour -- 95% marginal tax
1 day -- 90% marginal tax
3 months -- 50% marginal tax
1 year -- 15% marginal tax (today's capital gains rate)
10 years -- 10% marginal tax

I actually know a local guy who implements trading algorithms in programmable-gate-array hardware for the purposes of instant trading.

Re:Worthless (4, Interesting)

afidel (530433) | more than 3 years ago | (#35376734)

Nothing that crazy is needed just add a 10c per trade tax, the only problem is all the major trading countries would need to do it simultaneously or else the market would just move. I thought we were close a few years ago when the central banks and regulators started moving in lock step on policy but that only lasted until the first European debt crisis hit and China and the US got in the spat about currency valuations.

Re:Worthless (1)

Quiet_Desperation (858215) | more than 3 years ago | (#35376944)

Can we move the 15% out to 4 months? I'm just Joe Average, but I make some decent cash selling covered calls 2 to 3 months out. No need to pick on little old me.

Re:Worthless (2, Funny)

geekprime (969454) | more than 3 years ago | (#35376706)

A penny a share tax on any share traded will fix this problem immediately.

It will also dispose of all of the fraudulent "penny stocks" action.

It would also balance the US budget in about 8 hours.

Re:Worthless (0)

RightSaidFred99 (874576) | more than 3 years ago | (#35376802)

BS on multiple counts. First, there's no "problem". Second, obviously if it would raise that much in taxes nobody would do it, so it wouldn't balance any budgets.

Re:Worthless (1)

Anonymous Coward | more than 3 years ago | (#35376800)

The stock market is a system designed to encourage the optimal outcome of distribution for capital funding and the running of businesses, in exchange for which those who do so take on the risk of their investment offset by the prospects of future gain. Handing the optimization over of such processes to a computer is simply a way to compete with others.

I.E. If it served no purpose, why would anyone pay them? If it was so easy, why aren't you a billionaire?

Re:Worthless (1)

Nursie (632944) | more than 3 years ago | (#35376914)

Who says it's easy?

For a start you need a huge amount of capital to get started in HFT and then you need to be one of the favoured few. Look what happens when the little people start to figure out how they can join in the fun and predict the behaviour of other actors - they get charged with fraud (happened recently in sweden).

Something doesn't have to be easy to be worthless. Nor does it have to be worthwhile to humanity for it to be of utility to those speculating on it.

Re:Worthless (5, Informative)

Anonymous Coward | more than 3 years ago | (#35376918)

That's one of the arguments used by HF traders to justify their trade: they claim to provide liquidity to the market. Yes, this is true; however, HF trading violates one fundamental precept of the stock market that makes it work: that market forces determines the actual value of stock, based on an assessment of the value a company provides. There is no such valuation in HF algorithms (the concept of "value" is meaningless to a computer), only arbitrage. HF trading can and often does distort true market values. They can also cause a huge mess very quickly (as witnessed in recent years i.e. the Dow dropping a 1000 points in a very short time due to a chain reaction caused by algorithmic trades).

The ideal function of a stock market in an economy is primarily to raise capital so that wealth creation activities can occur. But because of human greed, a part of it will always merely function as a wealth distribution scheme. This is okay so long as the wealth creation activities are not impeded by the wealth distribution activities. HF trading shifts the balance toward the latter, and at some point, it can actually become a detriment to the economy.

Re:How is it not considered insider trading? (0)

Anonymous Coward | more than 3 years ago | (#35376884)

Seen Superman 4? It is the equal of Legalized worms. its o.k though because they help get the ones who write the laws elected

Worse than you think :P (5, Informative)

Anonymous Coward | more than 3 years ago | (#35376904)

And you also don't understand how deep this goes. This is not about trading fast. They are actually trading ahead of trade execution.

Flash trading is a practice in which some equity exchanges hold orders to buy and sell shares for a split second before making that information public (available to other exchanges). The exchanges' customers can view these prices ahead of other traders for a fee. High-speed computer software can take advantage of that brief period between when an order is placed and when it's executed to allow those members to potentially get better prices and profits by slipping in and making the trade themselves.

http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aZwoslIGa5JQ
http://www.marketswiki.com/mwiki/Flash_trade

Now, the time window is about 50-300ms that the orders to be executed are posted and the automated systems can intervene. Basically, if you have orders like following coming in within 200ms (1/5 of a second),

PUT 1000@31
PUT 500@30
PUT 500@30.1
CALL 1000@market
CALL 100@29.5

the flash orders will come in, buy the two sell orders and sell it @ 31 to the market order and you end up with,

PUT 1000@31
CALL 100@29.5

This effectively stole $950 from the market order. But then they will pay 2x the trade fees to the exchange to split in their trades ahead of the others. This isn't about "testing" the market, but simply going right in the middle between transactions and milking them for the most they can. It is not trading - it is stealing.

Re:Worthless (1, Insightful)

Antisyzygy (1495469) | more than 3 years ago | (#35376922)

You are simply wrong. Its just a more advanced form of a "merchant class". You may as well get pissed at your local grocer or furniture store. After all, you can go buy produce from a farmer, and hire someone to build you a couch.

Re:Worthless (1)

magarity (164372) | more than 3 years ago | (#35376924)

These people are parasites. They provide nothing of value to the world; they just take. This crap should be illegal.

The computer you used to write that rant was made possible by many, many investors who never even met combining their capital through the stock market to hire the management, hire the engineers, hire the construction workers, buy land and material for the factories, hire people to work in the factories etc, etc.

Re:Worthless (1, Insightful)

Nursie (632944) | more than 3 years ago | (#35376940)

The people who put up the initial capital, and who sold when the company had achieved real value?

Sure.

The parasites that suck money out of the system by interposing themselves between other trades at the millisecond scale?

Not so much.

Re:Worthless (1)

timeOday (582209) | more than 3 years ago | (#35376946)

But they're supplying all-important liquidity! Do you want to have to wait a whole *millisecond* to rebalance your retirement investments when Wall Street could reduce that to *picoseconds* for just a few hundred billion dollars per year?

light travels .3mm in a picosecond (4, Insightful)

PJ6 (1151747) | more than 3 years ago | (#35376518)

So unless they've found a way to break the light barrier, this is a load of bull.

Re:light travels .3mm in a picosecond (0)

NoseSocks (662467) | more than 3 years ago | (#35376550)

Mod Parent Up. This is a very important point. How long does it take to transmit a single packet's worth of electrons onto the wire? How long to get it over even a simple 1m length of fiber/cable? And how many packets are part of a given trade?
It's like measuring down to the thousandths of a mL using a standard graduated cylinder.

Re:light travels .3mm in a picosecond (1)

icebike (68054) | more than 3 years ago | (#35376662)

How long does it take to transmit a single packet's worth of electrons onto the wire?

Onto the wire?

Two computers sitting back to back with a fiber channel between them could not exchange enough data in that amount of time
to complete even one trade.

The article is nebulous at best, spun from careless choice of words of someone trying to impress.

I suspect, given the vast numbers of computers involved, and the total transaction count over the course of the trading day, it seems possible you might be able to do some hocus pocus math and divide total trades world wide by the time available and come to ridiculous numbers like these.

Many of these trades happen between one brokerage and another over private networks, or between two accounts in the same brokerage, probably in the same computer farm. Given enough such farms you can see that the sheer volume of trades could be astoundingly high.

None of these trades are settled in real time anyway, they are all carried in magnetic ink in computers. You can settle days later as long as you agree what the price was at the time the trade took place, so there is very little data to actually exchange in real-time. Representative trades are reported to the big boards just to give the market the sense of what is happening. Bulk trades at average prices are done after hours to even out the books and settle.

Re:light travels .3mm in a picosecond (2)

ko7 (1990064) | more than 3 years ago | (#35376844)

At 10 Gbps. each bit is 100 picoseconds apart. A single minimum sized Ethernet frame is 72 bits so there's 7,200 picoseconds just to get the smallest IPv4 frame possible onto a wire. At the speed of light in a vacuum (things travel slower on a wire or fiber) the first bit will have traveled less than 2.2 meters when the last bit gets sent. Speaking of computer trading in terms of picoseconds is nonsense. IMNSHO

Re:light travels .3mm in a picosecond (4, Funny)

Anonymous Coward | more than 3 years ago | (#35376556)

Moore's Law fixed that. Light can now travel 300 m in a picosecond. By the time this product is developed it should be able to travel a few km.

Re:light travels .3mm in a picosecond (4, Funny)

slyborg (524607) | more than 3 years ago | (#35376612)

You don't get it. The next step is Market On A Chip technology. The NASDAQ, NYSE, etc. will be condensed onto an integrated circuit in Lloyd Blankfein's office. But don't be concerned, the Market will still be FULLY FAIR AND TRANSPARENT for all...

Re:light travels .3mm in a picosecond (1)

Anonymous Coward | more than 3 years ago | (#35376938)

Fortunately the MOC will be equipped with a proper quantum coupled JTAG (QC-JTAG) interface to debug the markets. Just don't try to observe the markets for fairness or transparency as observations will change the state of the markets in an unpredictable way.

Re:light travels .3mm in a picosecond (0)

Anonymous Coward | more than 3 years ago | (#35376948)

"You don't get it. The next step is Market On A Chip technology. The NASDAQ, NYSE, etc. will be condensed onto an integrated circuit in Lloyd Blankfein's office."

No, it's YOU who do not get it.

The distance of 0.3mm precludes a lot of things, including that "Market On A Chip", unless of course, they can shrink that chip to a dimension much MUCH tinier than 0.3mm

Re:light travels .3mm in a picosecond (3, Insightful)

pz (113803) | more than 3 years ago | (#35376616)

I have a few friends in high finance. They're well-educated folk. So when they use certain terminology within the realm of finance, it makes them cringe because they know it's wrong, but the rest of the knuckle-draggers dont know the difference between a made-up Greek letter ("vega" ... no I'm being serious, look it up; it has to do with the volatility of options pricing) and a real one. When milliseconds are too slow, nanoseconds aren't a big enough improvement, they need to go one step beyond! No one with a brain is going to seriously consider speeding up by six orders of magnitude to the ludicrous level of picoseconds, but abuse of terminology is rife within the financial field.

Re:light travels .3mm in a picosecond (4, Informative)

mirix (1649853) | more than 3 years ago | (#35376810)

Nine orders of magnitude. 1ps is a billionth of a millisecond. (you forgot micro...). I know, brainfart, but 10^9 makes it that much more ridiculous.

Re:light travels .3mm in a picosecond (3, Funny)

zippthorne (748122) | more than 3 years ago | (#35376620)

It just means that the trading companies will be trying to co-locate.. to an adjacent core....

Re:light travels .3mm in a picosecond (2)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#35376728)

Why stop at the level of cores? Only grandmas and mutual funds let their algorithms languish in L3 cache, and Goldman has a 99 year lease on L1; but for just a little extra a few million gates of L2 can be yours...

Re:light travels .3mm in a picosecond (4, Interesting)

Kell Bengal (711123) | more than 3 years ago | (#35376654)

Well, this isn't a case of time-lag so much as it's a recognition that interactions with stock trading is actually a control system. The traders are trying to use feed-forward control to predict when and where the market is going to move and apply the right 'control action' (ie. buy or sell) at the appropriate time. Unlike most dynamic systems, however, you are not the only controller - you are trying to predict and exploit the behaviour of other controllers in the system.

With multiple players, the aggregate dynamics are something akin to a dog-fight as each trading algorithm circles and dodges, trying to exploit weaknesses and failures in its adversaries so as to make fractional gains in the time available. If you can control on a tighter time-scale than your opponents, then you can achieve more finely-grained dynamic buy-sell strategies that maximise your profit.

The fixed time-lag between you and the actual market is actually largely irrelevant because of the way the fast dynamic control strategy is being employed. To extend the metaphor, think of them like a attack formation flying to the enemy. If your aircraft can make their attack and withdraw before the enemy can reposition defenses, you will be more successful. The fact that it took an hour for your formation to reach the enemy in the first place is irrelevant.

Re:light travels .3mm in a picosecond (2)

countertrolling (1585477) | more than 3 years ago | (#35376908)

Don't they call that "leading your target"?

Re:light travels .3mm in a picosecond (1)

msauve (701917) | more than 3 years ago | (#35376764)

They skipped microseconds!

But, to your point, the latency (actually, queuing delay) of a minimal Ethernet packet at 10 gbps is ~51 ns at a minimum. Double, when you consider the return time, so a minimal trade would be on the order of 0.1 us plus processing time (that overclocked 5 GHz PC takes 0.2 ns per instruction cycle) on both ends (and add another 51 ns for the NIC to receive the frame in order to send it up a layer). And that only if you're within a few feet (~1 ns/foot) and on the same switch as the source data. So some sizeable fraction of a microsecond is a more realistic minimum for a select few who may have physical presence at the exchange. That's only 10000x difference from the BS in the article.

Re:light travels .3mm in a picosecond (1)

Anonymous Coward | more than 3 years ago | (#35376778)

The point is actually probably more....

My trade beat your trade by 3 clock cycles. Therefore, I get the better rate.

Since a clock cycle in these systems is on the order of a few dozen picoseconds, this is relevant.

Sure, you can't COMPLETE a trade in that time, but if you are 150 picoseconds faster than the other guy, you get the deal and he doesn't.

I AM just guessing here, but that's the only possible metric I can see any value in this measurement.

Femtoseconds (1)

i-linux123 (2003962) | more than 3 years ago | (#35376530)

I've started talking about femtoseconds.

Re:Femtoseconds (2)

falzer (224563) | more than 3 years ago | (#35376688)

Femtoseconds are so last microsecond. This just in: Trading geeks have started to talk about attoseconds!

Re:Femtoseconds (2)

zill (1690130) | more than 3 years ago | (#35376712)

You guys are too slow. I've re-implemented my trading platform in terms of planck time.

Re:Femtoseconds (1)

i-linux123 (2003962) | more than 3 years ago | (#35376744)

That's it, I am now booking a meeting with my team tomorrow to start talking about quantum teleportation. Touché!

Re:Femtoseconds (1)

Anonymous Coward | more than 3 years ago | (#35376792)

I willan on-hear your meeting and wioll-haventa onstolen your business plan. Thank you.

Wonderful (5, Insightful)

ArchieBunker (132337) | more than 3 years ago | (#35376532)

Thanks for fucking up the market for the rest of the world. This image comes to mind..

http://farm4.static.flickr.com/3014/2907411559_117ac480b5.jpg [flickr.com]

Re:Wonderful (-1)

Anonymous Coward | more than 3 years ago | (#35376588)

Butthurt and jealousy are ugly things. Y U SO MAD BRA?

Re:Wonderful (0)

Anonymous Coward | more than 3 years ago | (#35376594)

+1,000,000,000,000

But... (2)

jo42 (227475) | more than 3 years ago | (#35376534)

Wouldn't it be faster to just add some zeros to a number in a bank account? In the end, that is all that the modern stock market (AKA one giant ponzi scheme) does. Wall St. does sod all as far as producing real goods, real jobs and any real value.

Re:But... (0)

Anonymous Coward | more than 3 years ago | (#35376554)

Wall St. does sod all as far as producing real goods, real jobs and any real value.

Playing devil's advocate, the stock market basically does the very helpful job of putting funds into the hand of the most capable, most innovative companies.
I still think the whole system is stupid, though.

Re:But... (1)

kevinmenzel (1403457) | more than 3 years ago | (#35376586)

Yes, because a company can do great things with the funds that they have for a picosecond.

Re:But... (1)

jrumney (197329) | more than 3 years ago | (#35376750)

Perhaps that means the most capable, most innovative companies are not the ones trying to make money off picosecond trading.

Re:But... (2)

dakohli (1442929) | more than 3 years ago | (#35376668)

Now, I am not an expert. But, looking at the system, the only time a company receives money directly from an investor is during the IPO, or if they sell more stock. Most of the stock that is out there is already paid for. It is just moving around to make money for someone.

If I have over simplified it, please correct me.

Re:But... (1)

jrumney (197329) | more than 3 years ago | (#35376784)

Directly yes. But a higher share price still benefits a company indirectly. For example, a higher market valuation might give them an improved credit rating, giving access to more credit at a lower price.

Re:But... (1)

AuMatar (183847) | more than 3 years ago | (#35376832)

Nope, you're absolutely right. Unless the company makes an offering, none of the money on the market goes through their hands

Re:But... (1)

FooAtWFU (699187) | more than 3 years ago | (#35376628)

There are stupid bubbles. It's true. But in the long run the stock market actually does result in investment in things like factories and oil pipelines and other useful goods and services, and it works more effectively than if you had to finance it one tycoon at a time. The real Ponzi scheme is Social Security.

Re:But... (1)

timeOday (582209) | more than 3 years ago | (#35376932)

OK, go ahead and tell me who's the Bernie Madoff of Social Security. I'm waiting...

The Rich (0)

Anonymous Coward | more than 3 years ago | (#35376536)

And the rich get richer

impossible to process, utter rubbish (2)

rubycodez (864176) | more than 3 years ago | (#35376538)

light moves 0.3 millimeters in a pico-second. They are going to get all transactions on earth to occur in a sphere of 0.15 mm diameter? and somehow instantly get a traders transaction into that sphere from hundreds of kilometers away? That is pure inactionable bovine manure.

Re:impossible to process, utter rubbish (2)

masterwit (1800118) | more than 3 years ago | (#35376600)

Well I bet it is measured in the 10's of thousands. Reading your comment made me realize you have some insight there...found this one out:

[...]10^12 1 picosecond ps One trillionth of a one second 1 ps: half-life of a bottom quark
4 ps: Time to execute one machine cycle by an IBM Silicon-Germanium transistor 1 ps, 10 ps, 100 ps

10^9 1 nanosecond ns One billionth of one second 1 ns: Time to execute one machine cycle by a 1 GHz microprocessor
1 ns: Light travels 12 inches (30 cm) 1 ns, 10 ns, 100 ns

10^6 1 microsecond s One millionth of one second sometimes also abbreviated sec
1 s: Time to execute one machine cycle by an Intel 80186 microprocessor[...]
(source [wikipedia.org] )

I suppose the reasons they use picoseconds is primarily:
---at the scale of a microchip, 3mm is quite the distance
---the jump for units of measurement goes from a picosecond to a nanosecond. We all know nanoseconds could definitely be a bit slow in today's world and real decimal values are messy, unfortunately this means a jump from 10^6 to 10^9. We all know that is quite the jump in magnitude.

Re:impossible to process, utter rubbish (1)

Nemyst (1383049) | more than 3 years ago | (#35376702)

This is market trade we're speaking of. It's global. They're speaking of transactional processes on the level of a picosecond, but just reaching the central server HAS to take them at least a few miliseconds (it's the law!), excluding any possible processing.

It's a load of bull, seriously. Any scale below what it takes for light to travel a few dozens of kilometers is physically unfeasible unless we suddenly invent time travel or FTL micro jumps

Re:impossible to process, utter rubbish (1)

masterwit (1800118) | more than 3 years ago | (#35376740)

any possible processing.

Yeah, that is the only possibility of what I was thinking that could ever be considered... I guess the best way to look at this, if we are looking for a hint of truth, would be for the programming of very efficient algorithms to still produce calculations... after all calculations can take millions of pico seconds and a certain block of code executed 1 million times might be 50 picoseconds better than another per iteration. It all adds up internally.

Efficiency of algorithms is a big deal even with some latency. If you can react one second earlier than your competitors... that may mean a lot of money!

Yeah I would call bullshit though if they were referring to signal in the actual fiber.

Re:impossible to process, utter rubbish (1)

afidel (530433) | more than 3 years ago | (#35376786)

Nah, ping times in a HFT complex should be on the order of 100 nanoseconds (Infiniband port to port times). Even 10Gb Ethernet is only around 10us latency for short runs.

To put this in perspective (3, Informative)

masterwit (1800118) | more than 3 years ago | (#35376540)

To put this in perspective:
---A picosecond is roughly "330 picoseconds (approximately) – the time it takes a common 3.0 GHz computer CPU to add two integers" (source [wikipedia.org] )
---To put that in perspective, since obviously a large large amount of data must be inputted and then "processed" in real time, but then they are concerned with ~350 cpu cycles?
---Even if a processor can do tons of these operations a second, the amount of data they are receiving must be ghastly!
Makes me think of the patriot missile system and the round-off error tragedy that occurred. I am just hoping our market does not "experience the same fate". (I do understand it was all a fundamental bad programming situation before, but decisions that are made in picoseconds should be taken with some level of precaution.)

Re:To put this in perspective (2)

emurphy42 (631808) | more than 3 years ago | (#35376580)

TFA is slashdotted right now, so this is necessarily a guess, but maybe they're talking about e.g. 1,000 CPUs each doing one operation in 330ns as basically equivalent (in terms of net work done) to 1 CPU doing all 1,000 of those operations in 330ps apiece?

Re:To put this in perspective (1)

masterwit (1800118) | more than 3 years ago | (#35376638)

1,000 CPUs each doing one operation in 330ns as basically equivalent (in terms of net work done) to 1 CPU doing all 1,000 of those operations in 330ps apiece?

Haha, I was just saying that in reply to the comment [slashdot.org] I had posted as a reply above. 10^6 to 10^9 is the next jump in the verbal scale so even if they have to use 10,000 picoseconds as the basis, that is still better than dealing with the possibility of a real decimal representation in a nano-second. All the little guys add [slashdot.org] up after a while...

Re:To put this in perspective (2)

Thing 1 (178996) | more than 3 years ago | (#35376814)

[...] decisions that are made in picoseconds should be taken with some level of precaution.)

That just amuses me.

Real solution (0, Insightful)

Anonymous Coward | more than 3 years ago | (#35376542)

BAN short-term micro trading. It causes problems and offers no real benefit.

Re:Real solution (0)

Anonymous Coward | more than 3 years ago | (#35376848)

No! Markets become more efficient as time delays go to zero. The main issues with trading is deciding what constitutes a monopoly, insider trading or fraud. Also, you need to think of phantom assets as a type of fraud, which can be stopped with a low asset tax.

Hocus Pocus (5, Interesting)

bigmo (181402) | more than 3 years ago | (#35376552)

In Kurt Vonnegut's 1997 novel Hocus Pocus, the United States is brought to its knees financially by a company called Microsecond Arbitrage. Everyone invests through them and makes lots of money until a glitch happens and someone else's computer is faster that day. Then the entire country loses its shirt.

Word to the wise.....

Not good for the market: need synchronous clocking (4, Insightful)

Richard_J_N (631241) | more than 3 years ago | (#35376564)

Honestly, this is really a bad idea for overall market stability. What we really need is a much slower, yet fairer system.

What I'd suggest is something similar to synchronous clocking:
      Every second, on the second, prices are published.
      500 ms later, orders are placed and fulfilled.
      500 ms later, the updated prices are published.

Benefit #1: fairness - those who are closest to the exchange or have stupidly fast hardware can't get in front of the rest.
Benefit #2: slower responses. If the clock can only "tick" 60 times a minute, there is a chance for human intervention before disasters happen.
Benefit #3: markets are more able to serve the rest of society, rather than being used purely for "gambling". imho, the existence of "high frequency trading" is a kind of tragedy of the commons: nobody really "wins", but if everyone else does it, and you don't, you lose.

Re:Not good for the market: need synchronous clock (4, Insightful)

Nazlfrag (1035012) | more than 3 years ago | (#35376592)

I'd suggest every minute and 30 seconds respectively so human beings can also participate.

Re:Not good for the market: need synchronous clock (2)

earls (1367951) | more than 3 years ago | (#35376640)

And the next guy suggests an hour and half hours. The guy following disagrees, 500ms should be enough, 250ms to split the difference. I agree in principle that a line should be drawn, but how do you draw an arbitrary line that's fair and agreeable to all? I don't believe that's possible, hence you can never draw a line at all. YEEEEAH! HENCE!!

Re:Not good for the market: need synchronous clock (3, Informative)

Richard_J_N (631241) | more than 3 years ago | (#35376690)

Actually, I think we can draw a line. It takes about 200ms for an electrical pulse to travel round the world (speed of light in glass is lower than c), and we have a bit of switching delay. So this should imply the minimum timing limit. Anyway, fortunately the exchange can set the rules here, if it wants to.

Re:Not good for the market: need synchronous clock (1)

earls (1367951) | more than 3 years ago | (#35376726)

Meh, looks like it's irrelevant - the trade tax scenarios above seem to address the problem adequately.

Re:Not good for the market: need synchronous clock (1)

HornWumpus (783565) | more than 3 years ago | (#35376830)

It would be much better if the market solved this problem without government intervention and taxation. I doubt NYSE will or can. Goldman owns them. Another exchange could gain advantage by being trustworthy and level with all participants. It doesn't matter that the average investor doesn't play in day trading, much less anything shorter. Once you let the insiders openly tax the market nobody will ever trust them again.

The government already has too much money and still prints more at ever increasing rates. They should concentrate on fixing their own mess. Freddy and Fanny are undeniably government messes and still fester.

Re:Not good for the market: need synchronous clock (2)

earls (1367951) | more than 3 years ago | (#35376864)

People will go where the money is, any artificial limitation that minimizes profit margins will be ignored as long as more lucrative alternatives exist. I'm not sure how one would devise a self-sustaining system of integrity considering such.

Re:Not good for the market: need synchronous clock (1)

HornWumpus (783565) | more than 3 years ago | (#35376942)

It all depends on who's profit margins we're talking about. Traders trade, they generally don't buy or sell. Different participants etc.

It's not so much that Goldman Sachs can front load the day traders that bothers me (day traders are generally just gamblers/speculators, hence chumps). It's what else that access bought that brings the whole market into question. That can not be allowed to happen. All markets have skeletons in their closets and such snooping cannot be allowed. Even if it means Goldman has to give up a cash cow.

What I'm saying is perception is everything. If the general perception becomes that the NYSE is being skimmed by insiders many investors will go elsewhere.

If that loss of confidence were to somehow cascade into a loss of confidence in US government bonds then we would be done.

IMHO the only reason there is any interest in US government debt is an almost amazing lack of quality places to park capital. We owe the world a debt for being even more screwed up then we are.

Re:Not good for the market: need synchronous clock (0)

Anonymous Coward | more than 3 years ago | (#35376664)

Maybe once per day would be even better. Then we will all be able to see the train wreck in advance far enough that we can put a fix in before any real harm is done? (no sarcasm). Much like extending copyright to longer and longer periods, perhaps we have gone way too far with faster and faster trades and we need to seriously scale back to something that currently seems insane.

Re:Not good for the market: need synchronous clock (0)

Anonymous Coward | more than 3 years ago | (#35376636)

There is a country in sub Saharan Africa (honestly forget which) where the stock market is open every Tuesday from 10:00 to 11:00AM. The trading session consists of back and forth negotiations and ends with a single daily transaction that occurs at a single price that maximizes volume and does not leave any unfilled orders. Once a week. That's all you need.

Re:Not good for the market: need synchronous clock (1)

istartedi (132515) | more than 3 years ago | (#35376736)

This reminds me of the London gold price fixing [wikipedia.org] which seems irrelevant because gold is also electronicly traded, and probably HFT'd; but it looks like they still do it.

Re:Not good for the market: need synchronous clock (1)

HornWumpus (783565) | more than 3 years ago | (#35376868)

Once a week. That's all you need?

And as a reference you give us the economy of an unnamed southern African Nation? South Africa is the class of the lot and is nothing to envy.

How would that even work for an exchange that carried thousands of stocks?

What happens if 'shit happened' during the week. All the retail trades in a stock are ether buys or sells. Huge volume of done deals coming into the session and everybody knows it. Counter party holds their price and gets rich. Now retail has to deal with a week of trades at that price.

That extreme doesn't work ether.

Re:Not good for the market: need synchronous clock (0)

Anonymous Coward | more than 3 years ago | (#35376714)

That would be a MUCH bigger change than you think it is. All the financial institutions have billions invested in ultra-fast trading infrastructure (messaging software, switches, routers, fiber links, etc...). Not to mention all of the actual algorithms trading assume each quote modifies the market, which is immediately tradeable.

I kind of doubt the SEC would be able to enforce that regulation on exchanges and ATS's.

Re:Not good for the market: need synchronous clock (1)

Myopic (18616) | more than 3 years ago | (#35376716)

I like that idea, and I've thought of things like that before, but wouldn't there still be in incentive to be first in line with putting in your order? If everyone gets the prices published at a server in NYC at a particular moment, then the knowledge of that publication still propagates away from NYC at the speed of light. It's an improvement for the markets in some ways, but I don't think in that way.

Re:Not good for the market: need synchronous clock (1)

Richard_J_N (631241) | more than 3 years ago | (#35376748)

Well, you could put your order in any time before deadline, but orders are only executed at the instant of the deadline. So, you get all the "thinking" time you need, and once the deadline is reached, orders are processed simultaneously (or in random order). This is similar to how concert tickets (or, for that matter, new issues of stock) are sold: you have a week to post in your cheque, and then all envelopes are opened at the same time - in the interests of fairness, the should be shuffled first if there is a risk that not all orders can be fulfilled.

Speed of light fail (1)

jfern (115937) | more than 3 years ago | (#35376570)

Unless your computer is within 0.3 millimeters of their server, this isn't going to work. Any talk of latencies of less than a nanosecond is insanity.

In the not too distant future... (0)

Anonymous Coward | more than 3 years ago | (#35376574)

Speaking at a London conference on Tuesday, Donal Byrne, chief executive of Corvil, a high-speed trading technology company, caused a ripple of audible incredulity throughout the room when he suggested that trading speeds could be reduced to picoseconds in the not too distant future.

In the not too distant future it is possible that
Donal Byrne's nose could grow to tremendous lengths,
his pants could catch on fire,
and monkeys could fly out of his butt.

Re:In the not too distant future... (0)

Anonymous Coward | more than 3 years ago | (#35376678)

Speaking at a London conference on Tuesday, Donal Byrne, chief executive of Corvil, a high-speed trading technology company, caused a ripple of audible incredulity throughout the room when he suggested that trading speeds could be reduced to picoseconds in the not too distant future.

Unfortunately, Mr. Byrne's method of communication left those folks in the back of the room bereft of incredulity, as the people in the front were able to take advantage of the picoseconds' closer distance they were to Mr. Byrne.

Sales of awe and flabbergast were increased as a result.

This article is just (0)

Anonymous Coward | more than 3 years ago | (#35376602)

sensationalist bullcrap. A journalist trying to sell a non-story.

What this really means (0)

Anonymous Coward | more than 3 years ago | (#35376614)

The stock market is unstable. If stocks were based on real world value this would be a ridiculous concept.

End the Bullshit (0)

Tablizer (95088) | more than 3 years ago | (#35376644)

The stock exchange should put in a random delay around a few seconds on each bid to prevent trying to game the system and wasting resources chasing one-up-man-ship.

If a trade depends on a given price range, then include an optional min and max range. If the target price after the delay is out of the specified range, then the bid is canceled (although with a small processing fee).

Re:End the Bullshit (0)

Anonymous Coward | more than 3 years ago | (#35376710)

The stock exchange should put in a random delay around a few seconds on each bid to prevent trying to game the system and wasting resources chasing one-up-man-ship.

If a trade depends on a given price range, then include an optional min and max range. If the target price after the delay is out of the specified range, then the bid is canceled (although with a small processing fee).

Computers have no trouble at all taking randomness into account. It just makes the system more volatile.

Re:End the Bullshit (1)

Anonymous Coward | more than 3 years ago | (#35376732)

I've heard the argument on introducing a random delay, and I just don't buy it.

Investors (and traders) want certainty and introducing a random delay is the opposite of that. It means all market participants are structurally short a (very) short-dated option for which they are not compensated. Nobody benefits, and liquidity suffers.

Trading latency (3, Insightful)

woboyle (1044168) | more than 3 years ago | (#35376648)

I was a software developer of risk analysis tools for companies on the CBOE (Chicago Board Options Exchange). Milliseconds are significant when you need to hedge a position (balance your risks). Picoseconds? That is just idiotic, IMHO. Personally, I think we need to throttle back the trend toward automatic computerized trading. There are too many badly understood issues with regard to chaos effects in these time frames.

Re:Trading latency (1)

NoSig (1919688) | more than 3 years ago | (#35376718)

There is no amount of time too small to matter if what matters is being faster than someone else.

Unsolved Problems in Technological Society (2, Interesting)

Anonymous Coward | more than 3 years ago | (#35376686)

We still haven't solved the problem that was first noticed in the Industrial Revolution: How to occupy workers replaced by technology, and share the financial benefits of technology equitably.

Luddites and Communists attempted to supply answers early on.

Both answers have obvious flaws.

Later in the 20th century, it looked like Keynsian economics and moderate socialism might be the answer; but that's debatable because WW2 caused massive re-employment and reconstruction which occupied a generation.

Things seemed to be humming along in the late 20th century, the problem was forgotten--then the dot-com crash initiated what will most likely, in retrospect, be regarded as the true beginnning of a new economic and social crisis.

Consider the postal service--essentially frozen in hiring, trying to cut service, and headed for bankruptcy without government support. The replacement of mail with email is cited as a major reason. This is just one small example of technological unemployment.

What does this have to do with HFT? Well, HFT is one example of something the market creates in this situation. There's a general consensus that it isn't productive, and perhaps even harmful. Yet at the same time, it absorbes some of the otherwise unemployed.

When discussing this issue with a friend, he actually labeled me as a neo-luddite. I think that's unfair. I'm not saying we should perpetuate something like the postal service, just to maintain the status quo. Plainly, a policy like that could have negative long-term consequences, since the economy as a whole would be discouraged from innovating.

At the same time however, we still need to come up with an orderly way of compensating displaced workers, and preventing harmful "innovations" from arising in the wake of technological progress. The problem is, determining what is "harmful", who is "displaced", and what, if any "compensation" should be dispensed is fraught with political peril.

The problem remains unsolved and, IMHO, inadequately acknowledged by policy makers.

Re:Unsolved Problems in Technological Society (2)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#35376776)

Jay Gould is said to have remarked "I can hire one half of the working class to kill the other half."(unclear if he actually did; but it is a punchy line...)

With today's advances in robotics, we are likely to see an even more efficient solution to the problems of displaced workers and productive capacity in excess of purchasing power: Humans who are replaced by robots can simply be massacred by robotic-ally manufactured robots. Sure, this will eventually result in the replacement of the human race by a densely packed sphere of computronium around the NYSE, and a bunch of relentless hunter-killer drones; but how else are we going to achieve an infinite per-capita GDP?

Speed of light is not a problem (1)

NoSig (1919688) | more than 3 years ago | (#35376708)

Yes, light can only travel a tiny distance in 1 picosecond. You can still trade at that speed. It would require the traders to be running their programs on the same machine that is running the market. These programs would have to be implemented in silicon since 1/picosecond is a terahertz and we don't have general purpose terahertz machines - you'd even need more than a terahertz since a trade probably cannot be carried out in just 1 cycle. The real solution to this madness is to run the market at 1 hertz or less.

are we extinct yet? (0)

Anonymous Coward | more than 3 years ago | (#35376754)

can't happen soon enough.

This is where we'll find the first AIs (0)

Anonymous Coward | more than 3 years ago | (#35376760)

Come on, where else is so much effort being put into intelligent and prediction-capable computing? Mark my words, the first functional AI will come out of the stock market. That's actually a funny thought if you give it a minute, the world might not end when a sentient computer mainframe takes over the military systems of a major world power but instead when a sentient AI melts the financial system and plunges us into an economic dark age.

When the matchs burn. (0)

Anonymous Coward | more than 3 years ago | (#35376782)

Ivar Kreuger aka the "Match King" pretty much just made up most of the rules of today's business world as get-rich-quick schemes for himself.

Off Balance Sheet Entities
This means that details of an enterprise do not appear in the parent company's financial statements. Some of these entities were more or less secret. The associated debt, called "off balance sheet obligations", didn't appear in any financial statements of the companies Ivar controlled other than in summary form, if at all.[41] (In this context, it should be kept in mind that most of these financial statements were figments of the imagination in any case.) Albert D. Berning of the firm Ernst & Ernst, International Match's auditor, rationalized it at the shareholder's meeting in 1926. He said "it is only customary to consolidate the assets and liabilities of companies in such a balance sheet when a substantial majority of the outstanding shares are owned by the parent company. Where less than such a majority is owned, the shares are included as investments." [42] This invention gained rapid acceptance by others, e.g. Goldman Sachs and Lehman Brothers. The former issued 250 million dollars worth of complex securities (equivalent to about 3.75 billion in today's money) in 1929. Lehman issued similar obligations, which immediately rose 30 percent.[43] Enron used them extensively and in the recent financial crisis they played a major role in bringing down Bear Stearns and Lehman Brothers.

.3 mm (1)

mgrivich (1015787) | more than 3 years ago | (#35376798)

Light travels 0.3 mm in a picosecond. The article author doesn't understand what the words mean, and Mr. Bryne is somewhere between a scammer and an idiot.

You canna' change the laws of physics, Capt'n! (0)

Anonymous Coward | more than 3 years ago | (#35376822)

c = 30 cm / ns = 0.3 mm / ps

Microseconds, maybe. Picoseconds, though, is just taking trash.

Time Syncrhonization (0)

Anonymous Coward | more than 3 years ago | (#35376880)

Picoseconds likely is an exageration.

Current time synchronization requirements are currently 3s, moving down to 1s soon. Reconciling these trades likely is going to be quite difficult if anybody is trying to commit fraud...

The London Stock exchange was recently bragging that they can do trades in 12 micro seconds (12,000,000 bigger than 1ps).

The state of the art for network synchronization is on the order of 10ns. You can push that to hundreds of pico seconds if you have a really good GPS receiver and custom hardware.

Pushing it down to a nanosecond would require compensating for thermal effects. Light travels about a 30cm in copper in 1 nanosecond. In 1 picosecond it travels .3mm. This means that variation of the length of the cables connecting the systems trying to synchronize on the order of .3mm can changing the timing (introducing variation in the round trip times, which is what is the biggest factor in determining time synchronization).

That's not even getting into the interconnect technologies having latencies that are still well above the picosecond level.

I guess this is a long way of saying that the state of the art is several orders of magnitude away from nanosecond-level trading, let alone picosecond-level trading. Simple physics makes picosecond-level synchronization extremely difficult, and without synchronization, trading is going to be extremely difficult.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?