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!

JPMorgan Rolls Out FPGA Supercomputer

timothy posted more than 3 years ago | from the rejected-rejected-rejected dept.

Supercomputing 194

An anonymous reader writes "As heterogeneous computing starts to take off, JP Morgan have revealed they are using an FPGA based supercomputer to process risk on their credit portfolio. 'Prior to the implementation, JP Morgan would take eight hours to do a complete risk run, and an hour to run a present value, on its entire book. If anything went wrong with the analysis, there was no time to re-run it. It has now reduced that to about 238 seconds, with an FPGA time of 12 seconds.' Also mentioned is a Stanford talk given in May."

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

Ah, progress... (4, Funny)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#36727726)

It fills the heart with inspiration to watch the best and brightest constructing advanced computers to solve the problems of mankind...

Re:Ah, progress... (1)

Anonymous Coward | more than 3 years ago | (#36727946)

Someone needs to pay these scientists. Supercomputers don't get built with flowers and hemp. Eventually the tech will find uses in other fields, and we will have the greedy banks to thank for providing the funding for R&D.

Re:Ah, progress... (1)

Canberra Bob (763479) | more than 3 years ago | (#36728400)

Banks use their r&d for competitive advantage. The chances of them freely releasing their work to help humanity (and their competitors) is not very high.

Re:Ah, progress... (1)

KingAlanI (1270538) | more than 3 years ago | (#36729728)

like how some tech started out with military funding because that's where the money is?

Re:Ah, progress... (0)

Anonymous Coward | more than 3 years ago | (#36728120)

so they can pull a fast one on you even faster.

Re:Ah, progress... (1)

khallow (566160) | more than 3 years ago | (#36728256)

This is a special case of a trade problem. And trade, namely, deciding how to exchange valuable things with each other, is a big problem of mankind. So yes, my take is that it should fill your heart with inspiration.

Re:Ah, progress... (1)

Bengie (1121981) | more than 3 years ago | (#36728712)

you'll be glad to hear I have a 2TF super-computing chip crunching out massive frames-per-second as we read.

Objection. It's not math. (1)

schlameel (1017070) | more than 3 years ago | (#36728894)

They aren't doing math; they are using software. Patented, mathless software.

Re:Objection. It's not math. (2)

MachDelta (704883) | more than 3 years ago | (#36729402)

Mathless software?

That's like salt-free seasoning-salt, right?

Engineers solve problems (4, Insightful)

Weaselmancer (533834) | more than 3 years ago | (#36728918)

I would positively love to do something like this. The purpose of an engineer is to solve problems. That's what makes me happy at work. Solving problems. So here you have a very specific problem that required the construction of a custom computer made out of banks of FPGAs. Tell me that's not sexy! Who cares if it's for bankers. That is a damn nifty gadget to work on building.

Imagine building it yourself. Switching networks, Linux on ARM cores peppered here and there coordinating and dumping program code to the FPGA banks, writing the drivers to grab the data once the run is completed...

And at the end of the day a problem solved: What once took all day now takes a couple of minutes.

This would have been a thoroughly nifty machine to work on.

imagine helping JPMorgan destroy the economy (4, Insightful)

decora (1710862) | more than 3 years ago | (#36729194)

there have been articles about the computer guys who were working inside the CDO machines of wall street in 2000-2008 before the whole thing came crashing down.

the only people who could hold their nose and not-care what their work was being used for are complete pscyopaths, who went through some kind of personality-cracking process so that they can act like normal people while they help destroy the planets economy.

Re:Engineers solve problems (2, Interesting)

Anonymous Coward | more than 3 years ago | (#36729216)

Who cares if it's for bankers. That is a damn nifty gadget to work on building.

Replace "bankers" with "the Third Reich" and you would no doubt find some IBM engineers in the 1930s thinking exactly the same thing about the computers they were building.

Just because the technology is cool doesn't mean you're absolved of any moral or ethical dilemmas that may arise from working on it.

Speaking of scientists who worked with the Nazis.. (1)

KingAlanI (1270538) | more than 3 years ago | (#36729756)

Tom Lehrer's "Wernher von Braun" song, as one satirical example of science being divorced from ethics.

Re:Engineers solve problems (3, Informative)

hjf (703092) | more than 3 years ago | (#36729434)

Well, why don't you? The beauty if VHDL is that it's really, really basic. The rest is up to you to implement. Come join us in ##vhdl at irc.freenode.net or try fpga4fun.com. $100 should get you an FPGA training board which you can use to learn.

Re:Ah, progress... (1)

Alien Being (18488) | more than 3 years ago | (#36728948)

The brightest heartless are inspired by the best computers to advance the problems of mankind.

I eagerly await the next big bombing in NYC.

sick (1)

decora (1710862) | more than 3 years ago | (#36729202)

dude that is horrible

The US is different from Europe (0)

Anonymous Coward | more than 3 years ago | (#36729198)

Instead of going to Washington DC to run the country, the best and brightest go to Wall Street to run the country.

More math, faster... (5, Funny)

seanadams.com (463190) | more than 3 years ago | (#36727736)

That will fix our banking system for sure!

Re:More math, faster... (1)

blair1q (305137) | more than 3 years ago | (#36727766)

It just accidentally moved the decimal point on your account two places to the left.

But it did it hella fast.

Re:More math, faster... (1)

0100010001010011 (652467) | more than 3 years ago | (#36729514)

Ok! Ok! I must have, I must have put a decimal point in the wrong place
or something. Shit. I always do that. I always mess up some mundane
detail.

Absolutely! More golfing, less bankrupting! (2)

zooblethorpe (686757) | more than 3 years ago | (#36727844)

That will fix our banking system for sure!

I can't agree more. An increase in the number of golf matches and accelerated round-robin tournament configuration would go a long way to keeping those pesky bank officials occupied on the links and safely out of their offices. The Florida Professional Golfers' Association will benefit while also helping protect our nation's assets from executive malfeasance.

...

Wait a minute, what are we talking about here?

Cheers,

Re:Absolutely! More golfing, less bankrupting! (0)

Anonymous Coward | more than 3 years ago | (#36728526)

It may surprise you to learn that JPMorgan doesn't actually sponsor any golf events. Jamie Dimon has made a habit of eliminating golf sponsorship dollars at every company he's run.

But then, Slashdot comments aren't about accuracy, they're about mod points.

Re:Absolutely! More golfing, less bankrupting! (2)

KingAlanI (1270538) | more than 3 years ago | (#36729766)

I think it was basically a pun on another meaning of the FPGA acronym.

Re:More math, faster... (1)

Idbar (1034346) | more than 3 years ago | (#36728248)

Well we all hope that at least their bubble crashes faster, so they get a bailout and we all move on sooner.

Re:More math, faster... (1)

syousef (465911) | more than 3 years ago | (#36728864)

That will fix our banking system for sure!

Banks for sure?

Re:More math, faster... (2)

c0lo (1497653) | more than 3 years ago | (#36729186)

That will fix our banking system for sure!

GIGO at tremendous speed.

The best and brightest (1)

toppavak (943659) | more than 3 years ago | (#36727742)

A friend in the industry once remarked that some of the best and brightest in software engineering have been going into the financial industry as of late. It's hard not to wonder what they might have achieved in more productive areas of work...

Re:The best and brightest (0)

Anonymous Coward | more than 3 years ago | (#36728164)

Like advertising?

Re:The best and brightest (1)

That Guy From Mrktng (2274712) | more than 3 years ago | (#36729832)

Why would advertising or marketing need bright people when theres actually no challenge in the field, I mean.. HIT THA MONKEY iN TEH BELLAH AND WIN 9000 iPads. == thousands of hits.

Advertising is 90% targeting, if you are not getting any "smart" Ads, probably you are not the target for those or use adblock since you were getting lots of dumb Ads.

Weep, USA, weep! (2, Insightful)

Anonymous Coward | more than 3 years ago | (#36728708)

This technological breakthrough is an important milestone on the downfall of the mighty USA. When the brightest of a country are engaged into a completely nonproductive activity and wastes important resources (in terms of education, knowleddge and to some extent money) to achieve nothing which benefits the society, it's a high-mark - it's all downhill from here.

Let's come back here after 20 years and see how this comment stands up to time.

Weep, USA.

Re:The best and brightest (2)

dakameleon (1126377) | more than 3 years ago | (#36728730)

For me, it was pretty straightforward: Money, and opportunity. Software companies here in Australia tend not to gamble so much on fresh graduates, so the majority of opportunities that were attractive financially were in the consulting (Accenture, IBM, Infosys) or Financial space. As a poor student coming out of university, when I'm given the offers of $45k/yr vs. $60k/yr, plus bonuses, it's a pretty easy decision. Work hours tend not to be onerous, workplace conditions good, and the majority are in the main CBD. After a couple of years, experience and recruitment policies tend to railroad you into the industry unless you take a major step sideways.

That said, I'm always curious what "more productive areas" of work people consider to be out there for the average software engineer.

hahaha (1)

decora (1710862) | more than 3 years ago | (#36729210)

there are people who live off the difference in those salaries.

what did you buy with that exra fifteen grand? did you need it?

Re:hahaha (2)

Nursie (632944) | more than 3 years ago | (#36729570)

Not that this justifies it, or that I disagree with you, but here in Australia the cost of living is vastly higher than the united states.

The US and AU dollars are at rough parity, but it goes a hell of a lot further in the US.

Huh? (1)

PJ6 (1151747) | more than 3 years ago | (#36727746)

The company also supports Excel and all different versions of Linux.

Riiiight.

Re:Huh? (1)

dakameleon (1126377) | more than 3 years ago | (#36728738)

Not that hard. Excel, front of house. Linux, back of house.

Pictures of the rig (-1, Troll)

dotpotttt (2366294) | more than 3 years ago | (#36727762)

Looks really impressive [thoughts.com]

Re:Pictures of the rig (0)

Anonymous Coward | more than 3 years ago | (#36727772)

Goatse Alert!!

Re:Pictures of the rig (3, Informative)

blair1q (305137) | more than 3 years ago | (#36727780)

Do not click. Some sort of phony coprophile spam.

Re:Pictures of the rig (1)

jittles (1613415) | more than 3 years ago | (#36727896)

I thought there used to be a way to report spam?

Re:Pictures of the rig (0)

Anonymous Coward | more than 3 years ago | (#36727784)

I hate you.

Garbage in, Garbage out (2)

ka9dgx (72702) | more than 3 years ago | (#36727792)

Information asymmetry makes even the fastest analysis on the planet irrelevant as the data input is garbage. It is this lack of transparency which resulted in the housing bubble, etc.

The "too big to fail" banks regularly hide data from customers, regulators, and other branches of their own organization.

This is interesting because of the speedup of FPGAs, but don't be fooled by a second that it addresses an actual business need, other than PR.

Re:Garbage in, Garbage out (1)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#36727884)

It is rather curious that a company of JP Morgan's size would choose to address risk with computers rather than lobbyists. It has already been demonstrated that enterprises who matter can simply play "heads I win, tails you lose" with the public purse.

Re:Garbage in, Garbage out (1)

Anonymous Coward | more than 3 years ago | (#36728208)

Nah, but you need the computers. If you simulate the Risk real closely, estimating your max exposure, this means you have have more room to continue trading before higher ups have to sign off on getting more collateral to cover the current deals. If you wanted to skip using computers, you could just say "Meh, the rules say if I'm blind idiot, I have to put up, oh, 20% collateral. On the other hand, if I do a full evaluation according to these rules, the simulations state I only need, 14%. Score! I can do another... 3-4 million worth of deals with current limits." The traders love this stuff.

Ideally, though, it help for some trading that if some VP acknowledges that frankly you can't price some products so maybe those are product types your bank should maybe avoid. And not by, oh, handling the issue of correlation of deals (Ie, if these deals become riskier, do these other deals also become riskier?) by just declaring the correlation constant is 0. Which is perhaps fine when dealing with small portions of spread out markets, and less fine when you're actually starting to deal with a large percentage of the market at the same time.

Re:Garbage in, Garbage out (1)

MichaelKristopeit501 (2018074) | more than 3 years ago | (#36727902)

and the PR is basically an admission that their system architects are unable to build a report cache system that continuously regenerates.

Re:Garbage in, Garbage out (1)

khallow (566160) | more than 3 years ago | (#36728698)

Information asymmetry makes even the fastest analysis on the planet irrelevant as the data input is garbage.

Even garbage data tends to have real information hidden in it. And most financial data tends to have a quite a bit of real information in it.

Plus, what side of the "information asymmetry" is JP Morgan on? I bet they tend to have the advantage.

Re:Garbage in, Garbage out (1)

dakameleon (1126377) | more than 3 years ago | (#36728750)

Human stupidity, greed and arrogance resulted in the housing bubble & subsequent crash. Banks such as Goldman Sachs and Deutsche Bank who had clever people that saw the whole thing coming got out while the going was good and made money on the way down - powered by intelligence, greed and arrogance. Transparency had nothing to do with it.

Deutsche and Goldman almost bit the dust (3, Insightful)

decora (1710862) | more than 3 years ago | (#36729230)

if Morgan Stanley hadn't got bailed out by a Japanese bank, and if Bank of America hadn't bought Merrill, then Merrill would have failed, the Morgan, and Goldman and JPM would have fallen because of it.

Goldman's credit default swap business with AIG was also basically 100% bailed out by the taxpayer. Goldman would have lost massive amounts of money if it hadn't been for the deal the government gave them when it took over AIG.

.

A decade ago called... (1)

SmurfButcher Bob (313810) | more than 3 years ago | (#36727816)

...starbridge wants their concept back.

Needs Moar Technobabble... (1)

benjamindees (441808) | more than 3 years ago | (#36727832)

For the new Maxeler system, it flattened the C++ code down to a Java code.

Re:Needs Moar Technobabble... (1)

lennier (44736) | more than 3 years ago | (#36729484)

For the new Maxeler system, it flattened the C++ code down to a Java code.

Just one Java code? I bet it was '.'

I'm all tingly. (1)

musth (901919) | more than 3 years ago | (#36727840)

They can afford this crap because of all the money that has been transferred upward into bankers' pockets.

Heterogeneous Computing (0)

Anonymous Coward | more than 3 years ago | (#36727862)

I'm going to admit I didn't know what that term meant, and decided to google it. If you don't know I'd recommend you do the same, it looks very interesting, even if youj take nothing else from this story.

Hardware is faster than softare (1)

Smallpond (221300) | more than 3 years ago | (#36727958)

This is a good idea. A hardware [austinchronicle.com] implementation of a risk analysis algorithm is always faster than software.

What's next for the FPGA supercomputer? (1)

psyclone (187154) | more than 3 years ago | (#36727978)

Reprogram this risk analysis computer into a Bitcoin miner and finish the remainder of the 26M BTC?

How fast would a bitcoin server process the bank.. (0)

Anonymous Coward | more than 3 years ago | (#36728036)

I don't think the FPGA's are as fast as the ATI GPU's for computing SHA hashes, but they probably use a heck of a lot less electricity. What I want to know is how fast my 5 GHash bitcoin cluster would compute the banks book :)

Re:How fast would a bitcoin server process the ban (2)

HateBreeder (656491) | more than 3 years ago | (#36728272)

1 Virtex-6 SX475T could give you about 1 billion SHA-256 hashes/second clocked at 200MHz., will use 20% the power of the ATI GPU. but will cost about 4 times as much.

Re:How fast would a bitcoin server process the ban (1)

zill (1690130) | more than 3 years ago | (#36728444)

I don't think the FPGA's are as fast as the ATI GPU's for computing SHA hashes

Correct.

but they probably use a heck of a lot less electricity.

Correct.


But unfortunately their initial cost is much higher than ATI GPUs which is why no one started mass mining with FPGAs yet.

Re:How fast would a bitcoin server process the ban (0)

Anonymous Coward | more than 3 years ago | (#36728522)

I happen to know of someone who has been bitcoin mining with a huge array of FPGAs for the last two years and absolutely raking it in.

Re:What's next for the FPGA supercomputer? (1)

zill (1690130) | more than 3 years ago | (#36728488)

Actually it's 21 million BTC in total.

There are currently 14.2 million BTC left, which are worth $203 million USD in today's market values. JPmorgan made $17.37 billion last year. You do the math.

Common function migrates to hardware Film at 11 (0)

Anonymous Coward | more than 3 years ago | (#36728012)

OK, it is *somewhat* newsworthy, but in some sense it's not. Common function alwasy migrate into hardware. At some point, we were writing code to find the cos of an angle, then it got built into some languages, now it's built into hardware. Given enough time, the more common functions end up in hardware. Some day, most of the OS will be in there, and then all the stuff about Linux vs. $whatever will be pointless.

Have they got the right algorithm though? (1)

GumphMaster (772693) | more than 3 years ago | (#36728016)

So they have sped up to computation but have they spent any time and effort on getting a sane algorithm that, for example, can identify a house-of-cards situation like the recent collapse?

Re:Have they got the right algorithm though? (1)

DalDei (1032670) | more than 3 years ago | (#36728146)

They dont need to. As long as their algorithm is executed faster then their competitors they will win. Financial algorithms are not about sanity, they are about out-guessing the competition and making changes to your algorithm faster then they do.

Re:Have they got the right algorithm though? (1)

GumphMaster (772693) | more than 3 years ago | (#36728228)

They are modelling their risk exposure. If they have failed to adapt to recent risk stupidities then all they have achieved is a faster misrepresentation of their position. They might even get to ruin faster than their competitors, which is hardly a good thing even on a trading "floor".

Could they not use GPUs? (0)

Anonymous Coward | more than 3 years ago | (#36728020)

The article is light on details. It seems unlikely that they're doing anything that would prevent them from doing it significantly faster with less power draw on a GPU, and with lower initial hardware costs to boot.

Re:Could they not use GPUs? (2, Informative)

Anonymous Coward | more than 3 years ago | (#36728116)

It seems unlikely that they're doing anything that would prevent them from doing it significantly faster with less power draw on a GPU, and with lower initial hardware costs to boot.

No, FPGAs use significantly less power and provide greater performance than GPUs. The initial capital cost is higher though. Here's an article that gives a bit more detail: http://www.xilinx.com/publications/archives/xcell/issue74/FPGAs-speed-computation-complex-credit-derivatives.pdf [xilinx.com]

Re:Could they not use GPUs? (0)

Anonymous Coward | more than 3 years ago | (#36728518)

That article compares FPGAs to CPUs, where that claim is true, not GPUs. It's telling that GPUs are never even mentioned.

Re:Could they not use GPUs? (1)

gatkinso (15975) | more than 3 years ago | (#36728670)

>> No, FPGAs use significantly less power

Depends on the actual hardware being compared. The important metric isn't raw power consumption, but FLOP/watt.

>> and provide greater performance than GPUs

Not if the algorithm can be (and is!) expressed in kernels that can be operated in a massively parallel fashion. Not even close.

However if you just want to run Plain Old Code... then yes the FPGA is going to dominate. Also if you are looking for super low latency, then FPGA is the way to go.

I guess my point is that you can't make such a blanket statement.

Re:Could they not use GPUs? (1)

HateBreeder (656491) | more than 3 years ago | (#36728360)

GPUs are much more power hungry compared to FPGA and provide a fraction of the performance.

At the end of the day, GPUs are designed for gaming machines... the whole GPGPU thing is a side show for the graphics market. It's just not optimized in any way for this sort of computation. There's little money to be made building supercomputers compared to selling gaming machines.

However, an FPGA can be completely customized to suit your exact needs, you will make efficient use of the entire chip. It won't be a mere coincidence (like in the GPU case) that the chip can be used for a computation that you need. The FPGA is customized directly to fit an algorithm. this efficiency is where the speed gains are made.

It seems people put a lot of effort in to making their software compatible with GPUs and changing their algorithms to fit the GPU model.. this is a distorted view of reality - it is the computer that needs and can change to suit the problem, not the other way around.

Re:Could they not use GPUs? (0)

Anonymous Coward | more than 3 years ago | (#36729856)

GPUs from five years ago, maybe. Modern "graphics" processing units are streaming multiprocessors that are much more flexible than you seem to give them credit for. They also have much higher computational density, memory bandwidth, and computation-to-power ratios than FPGAs. The only thing an FPGA buys you is flexibility to solve problems GPUs are not good at, which they don't appear to need.

You know you have too much $$$ when... (0)

Anonymous Coward | more than 3 years ago | (#36728026)

... you need a super computer to add it up for you.

JP Morgue - Buying off Congress, taking hostage the commodity sector, buying off the CFTC, getting multi-billion bail-outs and free Fed POMO $$$ each day risk free! WTF do they need super computers to assess risk for? All they need is a macro to do what they do now - get the American taxpayers to buy their toxic assets they've offloaded to the Fed.

Re:You know you have too much $$$ when... (1)

nurb432 (527695) | more than 3 years ago | (#36728472)

And if you are real good you get to do the books for disaster area.

im less impressed by the (2)

nimbius (983462) | more than 3 years ago | (#36728118)

performance of the system, its architecture or its value to the company
and more impressed that one of the worlds largest financial service providers
partially responsible for the worlds second largest economic collapse has found, despite their
prior record with the concept of 'risk', the objective, quantifiable definition of the amorphous and
highly elusive concept of said 'risk.'

Re:im less impressed by the (0)

Anonymous Coward | more than 3 years ago | (#36729600)

Well, look at it this way, they now have the time to run MANY MANY different versions of risk in the previous span of time and compare them to what really happens!

Pretty stupid approach. (0)

RightSaidFred99 (874576) | more than 3 years ago | (#36728184)

Way to tie yourselves to some oddball implementation forever. In 5 years they'll be regretting the hell out of this decision. They should have invested the time and money in improving the algorithm and implementation, and thrown commodity hardware at the problem. Hardware will improve over time and it becomes an off the shelf improvement, not some horrible one-off solution they're stuck with barring great expense and yet another complete reworking.

The company that did this for them, however, was god damn smart. Guaranteed revenue to develop, support, and migrate off of this over time.

This is why the market crashed - these "smartest guys in the room" aren't so smart sometimes.

Re:Pretty stupid approach. (3, Insightful)

yodleboy (982200) | more than 3 years ago | (#36728252)

well as someone earlier posted, it's about making decisions faster than the competition. If they can analyze a transaction, assign a risk and make a decision faster, they make the money. If the competition also takes 8 hours to do an analysis, doing it in 3 min is huge. Who cares if it's outdated in 5 years? You build another with a fraction of the profits this one helped you earn you and keep chugging along.

sometimes, in some industries, first post IS important.

Re:Pretty stupid approach. (0)

Anonymous Coward | more than 3 years ago | (#36728330)

Yeah, but I can't believe they couldn't do this with a more mainstream approach. Besides, 5 years was generous. Between GPGPU and just plain old CPU improvements, more than likely like 1.5-2 years. They're not dealing with climate change or nuclear explosions - they don't need a god damn custom supercomputer. I think it's like those assholes who buy $50k golden plasmas that have the same picture as a $5k plasma, they just have money to burn.

Re:Pretty stupid approach. (1)

HateBreeder (656491) | more than 3 years ago | (#36728410)

I think you're underestimating the bank.

The cost of this solution might have been low enough to warrant the immediate gains in performance.
The lock-in you describe might not exist, as the algorithms and the accelerated bits are a small portion of the entire code-base (but take 99% of the run-time).
It will very likely be the case that the cost of not going with this solution is far far greater than going for it.

Re:Pretty stupid approach. (0)

smellotron (1039250) | more than 3 years ago | (#36728616)

In 5 years they'll be regretting the hell out of this decision. They should have invested the time and money in improving the algorithm and implementation, and thrown commodity hardware at the problem....
This is why the market crashed - these "smartest guys in the room" aren't so smart sometimes.

This statement appears to be make two assumptions:

  • Anyone who doesn't use commodity hardware is not smart
  • The people who make software implementation decisions at JPM also make the trading and investment decisions.

The first one is just ridiculous, and you should know it. Specialized hardware doesn't exist for shits and giggles, and JPM is big enough to understand both the maintenance cost of specialization and the performance gains. The second assumption... well... it's much more likely to be a "Death Star Contractor" situation. I doubt the head traders and investment officers at JPM really care about FPGAs, and I doubt that the R&D lab understands all of the nuances of the derivatives trading in which their company takes part.

Re:Pretty stupid approach. (0)

Anonymous Coward | more than 3 years ago | (#36728820)

> The people who make software implementation decisions at JPM also make the trading and investment decisions.

The people who make the software implementation decision at JPM sit alongside the traders making the trading decisions. Both sides know precisely the benefits from faster risk calculations.

Re:Pretty stupid approach. (1)

jamesh (87723) | more than 3 years ago | (#36728700)

and thrown commodity hardware at the problem. Hardware will improve over time and it becomes an off the shelf improvement, not some horrible one-off solution they're stuck with barring great expense and yet another complete reworking.

... because FPGA's are highly specialised one-off hardware items that are never improved on and will be obsolete within a year?

Re:Pretty stupid approach. (1)

WorBlux (1751716) | more than 3 years ago | (#36728714)

FGPA's are commodity hardware. No reason to believe they also won't improve in the next five years, even more so than a conventional CPU. as nobody can really see a way to get the manufacturing tech down below 14nm.

Re:Pretty stupid approach. (1)

Digicrat (973598) | more than 3 years ago | (#36729284)

Just because they're FPGAs doesn't mean they can't be upgraded over time. Most FPGAs can be re-programmed many times in the event that they improve upon their algorithm. Further, while FPGAs are improved upon at a slower rate, they are improved over time and I would suspect that their design could be ported to newer FPGA versions as they become available without to much trouble - just the expense of buying a new system. And given how much money the banks have, and the value they seem to be placing in this system, I don't doubt that they'd have any hesitation about purchasing a second "super-pipelining-machine" that's "even better" a few years down the line.

Re:Pretty stupid approach. (1)

hjf (703092) | more than 3 years ago | (#36729510)

No. ALL FPGAs can be reprogrammed. Field-Programmable gate arrays.

FPGAs don't have programming memory, they need an external chip that holds the program, and downloads it to the FPGA on boot. Ignore OP, he doesn't know shit.

Re:Pretty stupid approach. (0)

Anonymous Coward | more than 3 years ago | (#36729394)

Stealing from JP Morgan - that alone ought to be worth a drink in some places.

Maxeler's partially owned by JP Morgan (2)

ecner (2366434) | more than 3 years ago | (#36728202)

Note that Maxeler sold 20% of itself to JP Morgan earlier this year.

Nifty, but remember: (1)

Hartree (191324) | more than 3 years ago | (#36728204)

A purpose built machine will often vastly outperform a general purpose computer.

The nice thing about an FPGA machine is that creating the purpose built machine is much faster.

Risk and "Risk" (1)

Sponge Bath (413667) | more than 3 years ago | (#36728222)

Since they closed their proprietary trading operation, this will no longer be useful in evaluating bets against the "investments" they recommend to their customers. Is this really useful, or is it just another complicated black box they can show marks and say "this is too complicated for you to understand, so just trust us."?

Re:Risk and "Risk" (1)

dakameleon (1126377) | more than 3 years ago | (#36728804)

The result of this will be that they'll get a number summing up their credit risk in 5 minutes instead of 8 hours. If there's anything way out of whack, they'll be able to react during the day, rather than having to react effectively 24 hours later.

From the article.... (1)

gatkinso (15975) | more than 3 years ago | (#36728358)

"For the new Maxeler system, it flattened the C++ code down to a Java code."

???

Re:From the article.... (1)

MacGyver2210 (1053110) | more than 3 years ago | (#36729076)

By "flattened" it meant "bloated through interpretation and unnecessary recompilation".

C++ to Java? (2)

zill (1690130) | more than 3 years ago | (#36728382)

For the new Maxeler system, it flattened the C++ code down to a Java code.

I hope to God that's a typo. C++ -> Java -> Java Bytecode -> Native code almost sounds like a programming language Rube Goldberg machine.

How would the pointer operations even translate?

Re:C++ to Java? (5, Informative)

Anonymous Coward | more than 3 years ago | (#36728516)

That's not how it works. The Maxeler FPGA compiler is actually a Java library for describing dataflow machines. They call it "stream computing" - basically just very deep pipelines. You describe your algorithm in the dataflow/streaming style, using the Maxeler Java library, then when you run your program the output is an HDL design to load onto the FPGAs. There's no automatic conversion from a serial program (in C++ or anything else) to a dataflow/streaming program. The developer reimplements the algorithm manually.

Re:C++ to Java? (1)

bunratty (545641) | more than 3 years ago | (#36728532)

Java references are pointers.

Re:C++ to Java? (1)

renrutal (872592) | more than 3 years ago | (#36729894)

Usually when one asks if language X has pointers, the person wants to know if it can do pointer arithmetic, and that's a resound NO in JVM based languages. Which is why we say Java has references, not pointers.

Baloney (2)

bill_mcgonigle (4333) | more than 3 years ago | (#36728392)

The devs are going to run jobs while the machine is idle to corner the Bitcoin market.

But, wow, from the perspective of getting the Boss to buy awesome hardware for your pet projects - hey, we're not worthy.

Re:Baloney (0)

Anonymous Coward | more than 3 years ago | (#36728962)

Obviously, you know nothing about the issue of doing monte-carlo simulations on a portfolio full of derivatives, if you think this is baloney. Industry is flirting with GPGPU and FPGA for a while, cause in 2005 you already had cluster of 3000 machines (for instance http://www.banktech.com/news/showArticle.jhtml?articleID=171203551), and now banks run cluster of 10th of thousands of cores routinely.

Btw, if a dev for a large investment bank best idea of idle hardware use is to mine bitcoins, then he is seriously retarded.

Their risk metrics are less than worthless (2)

decora (1710862) | more than 3 years ago | (#36729174)

JP Morgan, along with every other mega-bank, has no idea what is actually on it's balance sheet, and hasn't for 10-20 years.

The Shadow Banking system is too big, too complicated, and too interconnected for any of these risk metrics to mean anything.

JP Morgan did the same business as Goldman Sachs and the others, loading up with CDOs and credit default swaps and CLOs and the rest of it. JPM is portrayed as 'wiser' than the rest in the books and the articles about the crisis, but its not really true. They were less stupid than the stupidest people, that doesnt make them smart.

They got bailed out just like all the other megabanks. Why? Because they had no idea what was on their books. They are running a black box. All the supercomputers in the world cannot make up for a complete and utter lack of transparency. And that is what the world of Credit Default Swaps (invented at JPM no less) are. A gigantic black box. The rest of the Shadow Banking system is the same, and JP Morgan (and the rest of the big banks) are up to their necks in it.

Just because you can calculate lies faster, doesnt mean they aren't lies.

Re:Their risk metrics are less than worthless (0)

Anonymous Coward | more than 3 years ago | (#36729348)

My experience of dealing with JPM was that their traders were absolutely insane but they made up for it by having the stubbornest legal department on the street. They were saved from themselves by the fact that it was just so much easier to get an ISDA with GS/MS/UBS/DB/...

RiskMetrics is a nice toy, but ultimately it's based on the same bogus historical vol fallacy as any other VaR framework.

The great tragedy of the last 25 years of finance is that there's no generally accepted Greek for liquidity.

Garbage In, Garbage Out.

The first step in the evolution of true AI... (0)

Anonymous Coward | more than 3 years ago | (#36729660)

This how I envision real artificial intelligence finally taking off. Sets of specially designed FPGAs continuously updated by other FPGAs which are specially designed to write code for the first FPGAs based on past the performance of both the FPGAs being programmed and other FPGAs in other sets as well as environmental input. Some FPGAs even rewriting their own code as necessary. All in a continuous round robin feedback system to constantly fine tune the operation of the entire system. After a few iterations, it may be difficult to determine how the whole thing works together even if we can do a dump of it all for detailed analysis, just as with a biological brain. The initial programming will be nothing more than the DNA to get the thing started. The basic desires and emotions, plus the basic mechanisms for reprogramming and adaptation. This, along with the hardware design itself and environmental conditions will determine how the FPGA cluster comes out in the end.

Overall, this may be a better mechanism than a biological brain because it will even be able to change the basic algorithms used for adaptation itself. Those algorithms in biological brains are only changed through millions of years of evolution, and haven't changed much for millions of years. (What has changed is the initial hardware construction mechanism. But still it is a slow process.) But they could be changed in seconds in an FPGA. The only thing that can't be changed - within the FPGA cluster by the FPGAs in that cluster - is the hardware itself. (I know FPGAs are, in a way, hardware that is changeable. However, in this context I mean changing which wires physically touch which wires and which FPGAs are plugged in where.) However, connect an FPGA cluster to a communication system which is connected to an automated FPGA cluster factory..... and well ....

All I can say is, you would soon be welcoming your new FPGA overlords.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?