![]() ![]() |
May 7 2010, 02:29 AM
Post
#1
|
|
|
The Dragon Never Sleeps ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Admin Posts: 6,924 Joined: 1-September 05 Member No.: 7,667 |
|
|
|
|
May 7 2010, 02:35 AM
Post
#2
|
|
|
Moving Target ![]() ![]() Group: Members Posts: 312 Joined: 3-March 10 Member No.: 18,237 |
Haha yeah, that was my first thought when I read about this...
|
|
|
|
May 7 2010, 02:45 AM
Post
#3
|
|
|
Moving Target ![]() ![]() Group: Members Posts: 565 Joined: 7-January 04 Member No.: 5,965 |
see, this is what happens when you let a hacker/technomancer get access to electronic stock trading. and the obvious reason to hire shadowrunners to pull this off, would be to then take that exact split second to buy buy buy at the artificially lower price. yeah, its a good run idea but you wanna hire prime runners from the new york sprawl.
hmm, if the OS for the NYSE went AI, things would get interesting in a hurry. |
|
|
|
May 7 2010, 03:21 AM
Post
#4
|
|
|
Immortal Elf ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 10,289 Joined: 2-October 08 Member No.: 16,392 |
Not only that, but I've heard that there's one stock trading software that specifically sells out to companies server rack space closer to their servers so that those companies have a microsecond edge on trading than other companies who's servers are on the other side of the room.
|
|
|
|
May 7 2010, 04:01 AM
Post
#5
|
|
|
Great Dragon ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 5,679 Joined: 19-September 09 Member No.: 17,652 |
Not only that, but I've heard that there's one stock trading software that specifically sells out to companies server rack space closer to their servers so that those companies have a microsecond edge on trading than other companies who's servers are on the other side of the room. That seems a bit unlikely. I mean, if company A is 0.000001 seconds closer to the server than Company B... well, would that really matter when Company A and Company B still have to send the signal to do whatever from locations X and Y? And really, given the human reaction time of somewhere closer to half a second or so, does 0.000001 seconds really make any difference? I'd be willing to bet they all have the same length cord connecting them anyway. |
|
|
|
May 7 2010, 05:13 AM
Post
#6
|
|
|
Moving Target ![]() ![]() Group: Members Posts: 821 Joined: 4-December 09 Member No.: 17,940 |
With such short time intervals, I'd think the algorithm's reaction time will be the determining factor - and slowing donw the server with more programs might lose the benefits.
Leasing an on-site server rather than relying on one a few kilometers away is going to be more significant as you dispense with the routing and get a faster and more reliable connexion Though in a shadowrun context I don't think many companies would be willing to place their in-house trading algorithms on a third party server - too much chance for spying or tampering with the transactions. |
|
|
|
May 7 2010, 12:01 PM
Post
#7
|
|
|
panda! ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 10,331 Joined: 8-March 02 From: north of central europe Member No.: 2,242 |
Though in a shadowrun context I don't think many companies would be willing to place their in-house trading algorithms on a third party server - too much chance for spying or tampering with the transactions. more like own servers in a third party location. Possibly something like a safe deposit box setup, where breaking into another corps rack would require some serious tools and no guards looking over the shoulder. |
|
|
|
May 7 2010, 12:32 PM
Post
#8
|
|
|
Runner ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 2,536 Joined: 13-July 09 Member No.: 17,389 |
That seems a bit unlikely. I mean, if company A is 0.000001 seconds closer to the server than Company B... well, would that really matter when Company A and Company B still have to send the signal to do whatever from locations X and Y? And really, given the human reaction time of somewhere closer to half a second or so, does 0.000001 seconds really make any difference? I'd be willing to bet they all have the same length cord connecting them anyway. In short yes. The long. That 0.000001 seconds does make a difference since a lot of trading is automated via triggers that a human creates. Let's say you bought a stock at $20 and you have a tolerance for 10% loss on that investment. That means when the stock price hits $18, you want to sell. Given that this was a matter of millions of stock (let's say 1,000,000 for the sake of argument) that was intended to be sold (instead of billions), each penny below $18 is a loss of $10,000 that you don't want. In the grand scheme it doesn't matter if you're the only person who is deciding to sell at $18, but if anyone else has a watch for $18, then you want to make sure you sell before he does, otherwise you will be selling below that $18. That's where the 0.000001 seconds come into play since all transactions are queued as a FIFO. If I get my transaction there at +0.001 seconds from when the stock price hit $18, and my opponent gets it there at +0.001001 seconds after it hits $18, I'll be sell at $18 and my opponent will be selling at a price below $18. Also, that distance traveled has a double effect. If it takes 0.00001 seconds for something to travel from the server to switch, then it will take 0.00002 seconds just to receive notice of a stock price change and send the sell/buy transaction. So if my travel time is only 0.000005 seconds instead of 0.00001 seconds, then I can send and receive in the same amount of time that it takes someone to receive. The servers will also over tuned to over perform to minimize the amount of time taking to load data into memory and process something in the CPU. I'm pretty sure if I saw the specs on these servers and the load they handle, I would cry. |
|
|
|
May 7 2010, 02:20 PM
Post
#9
|
|
|
Immortal Elf ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 10,289 Joined: 2-October 08 Member No.: 16,392 |
I'm pretty sure if I saw the specs on these servers and the load they handle, I would cry. Now you know why Deus took over the NYSE servers for a recompile. (IMG:style_emoticons/default/wink.gif) |
|
|
|
May 7 2010, 02:35 PM
Post
#10
|
|
|
panda! ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 10,331 Joined: 8-March 02 From: north of central europe Member No.: 2,242 |
iirc, much of the hardware used are GPUs from nvidia or ati/amd.
btw, anyone else worried about the attitudes presented in the comments on the article page? Or am i looking at a whole lot of teens trying to be ironic, or just trolling? |
|
|
|
May 7 2010, 03:13 PM
Post
#11
|
|
|
Immortal Elf ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Group: Members Posts: 10,289 Joined: 2-October 08 Member No.: 16,392 |
btw, anyone else worried about the attitudes presented in the comments on the article page? Or am i looking at a whole lot of teens trying to be ironic, or just trolling? I didn't even read the full article, so I didn't get that far. The first one made me laugh though: "It's all the democratic government's fault so they can pass the laws they want!" |
|
|
|
![]() ![]() |
|
Lo-Fi Version | Time is now: 21st June 2026 - 11:29 PM |
Topps, Inc has sole ownership of the names, logo, artwork, marks, photographs, sounds, audio, video and/or any proprietary material used in connection with the game Shadowrun. Topps, Inc has granted permission to the Dumpshock Forums to use such names, logos, artwork, marks and/or any proprietary materials for promotional and informational purposes on its website but does not endorse, and is not affiliated with the Dumpshock Forums in any official capacity whatsoever.