Australian (ASX) Stock Market Forum

Thinking of setting up a trading room

Anything but a fibre connection is a no go.

Rubbish.

m.

marklar I suck about 300 meg of data between the Aussie market hours. Times that by 20 traders + the ones that goof off download non-trading related Stuff and you have some serious data requirements. Get someone who wants to run an Algo Bot and they will be needing a stack more than what I pull in data.

And this has to be with very little latency. You would know more than me but do you think a ADSL+2 would cut it??
 
Every other decent shop has a proper connection, why allow others to get an edge over you? Have fun with your lag.
I used to work at Telstra designing, building & running Internet gateways and Data Centres for Federal Government & large corporates. I do know what I am doing.

m.
 
I used to work at Telstra designing, building & running Internet gateways and Data Centres for Federal Government & large corporates. I do know what I am doing.

m.

So you would say that the data requirements I posted above would be OK over ADSL?
 
I used to work at Telstra designing, building & running Internet gateways and Data Centres for Federal Government & large corporates. I do know what I am doing.

m.


Slightly different requirements. They don't need it INSTANTLY. Trust me, my parents are both in IT. My mother is a project manager in HealthSmart where they are updating servers. The requirement is not speed when you are transferring data. You have obviously seen how long an average sized Oracle DB works on a decent server. Running a bot has to be instant. With traders you can potentially be exchanging packages of a few GB a day. Lag will kill. I wouldn't risk it with ADSL if I was a day trader. For me though, I use a laptop and I don't need fast internet. For a day trader, like TH, I think ADSL will be a little slow.
 
So you would say that the data requirements I posted above would be OK over ADSL?
I never said it would. Doesn't need to be fibre though, there are other circuit connection types like Frame Relay to explore.

It could be done with ADSL, you could bundle multiple circuits together with something like this.
 
Not to mention having a sqawk-box running, multiple traders on youtube downloading interviews/bloomberg crap while watching Bernake/Trichet/Geither speak.

This all needs to be without lag.



You could of course avoid this problem and only take in traders who's average hold time is 30mins+.
 
I never said it would. Doesn't need to be fibre though, there are other circuit connection types like Frame Relay to explore.

It could be done with ADSL, you could bundle multiple circuits together with something like this.

Thats why I'm asking. I don't know. You would. What would be the best. Thinking that 10 GOOD traders will pull 1 to 5 g's a day each in profit so cost cutting a few dollars would not be the focus? Lowest latency connections with high reliability would.
 
Chap's

After you have sorted out your nuts & bolts, you need to find a provider for execution and leverage. Not retail leverage, but Prop. leverage, viz *100+

If any of you are actually any good, this will obviously be important to generating high dollar returns. When I say high, $100K/day are the amounts I've seen.

If you trade the US, there are numerous providers.

jog on
duc
 
Hey guys,

Would you guys use this? Any ideas? pros/cons?

I would greatly appreciate any feedback you can give me...At the moment, this is only something I am thinking of doing...

Interesting concept...the first thing to ask yourself is what need are you looking to fill, and who for?

Are you looking primarily to:
A) Generate Income - run it as a profit making business for traders:
B) Get access to / provide mentoring - an atmosphere to bounce ideas off others and learn from each other - trading can be lonely & boring at times:
C) Something else?

If A - you'd want to define what gap there currently is in the daily lives of your target market and establish what they would be prepared to pay (both to you and any associated costs) to fill it. For example,a lot of traders would need a damn good reason to shower/shave/dress and face a city commute to do exactly what they do each day from their living room in their jocks (Apologies if I have stirred up any unpleasant mental images) with no travel time - this is a personal cost in addition to what you would charge. A good trading library, professional speakers, courses, discounted data feeds, strategy reviews by peers and professionals - something they cant get at home (steady, nunthewiser!).

Bottom line is to look at this as a revenue generator youll have to research and plan it like any business (which is what you are starting to do by the looks of things) so youll have to create a business plan coving the 5 C's:

Company - Vision & mission, SWOT analysis,
Context - business environment and influential forces,
Collaborators - partners, financiers, suppliers etc - ,
Customers - segmented market profile - you are researching this one right now :)
Competitors- alternatives to what you will offer, eg stay at home

and 7 Ps
Products (define exactly what you will offer),
Price points (how much youll charge especially in comparison to alternatives),
Place (where you are operating eg Melbourne CBD serviced offices, suburban shopping mall, hired cybercafe),
Promotion (how you'll advertise and get ppl in the door)
People (Staff & Specialists),
Process (described in detail for any services),
Physical Evidence (referrals)

Once you've got this together youll have a solid idea of it it will or wont be able to make money.

If B - a value add in itself - it will be vvv expensive to hire and kit out office space in the city - maybe start small and start up, or join, a local traders group, get a bunch of people to commit to join you for a months trial, and then look for cybercafes in the area with good facilities and an understanding owner. See if you can come to an arrangement to book out as many machines as you need with dual screens at a negotiated rate between 9am and 4:30 PM for a four week block to start with - you should get a good discount with a guaranteed months income on weekdays, as long as you are not near a school or uni (hard core gamer territory). This would also be a good test if you are considering option A as well.

If C, please let us know!

Hope this is of some benefit.
 
marklar I suck about 300 meg of data between the Aussie market hours.

You must have much more demanding feeds than I have, as ruynning IB/TWS during the day takes no more than 80mb.

If any of you are actually any good, this will obviously be important to generating high dollar returns. When I say high, $100K/day are the amounts I've seen.

This sounds pretty ridiculous in the context of this thread. It sounds a gold digger in NYC suggesting 500k a year is poor.
 
Thats why I'm asking. I don't know. You would. What would be the best.
The simple answer is "it depends", there isn't enough detail here to propose anything, give me a full requirements spec & approximate budget and I'll see what might fit. As a general rule:
  • fast
  • cheap
  • reliable
pick any two.

Lowest latency connections with high reliability would.
Ok, well, it's going to cost you quite a bit. What is your budget? Can you codify your definition of "lowest latency"? (eg. average milliseconds round-trip time to where? U.S.A or Australia?) 300ms?, 100ms? per packet? can I tune your IP stack? What type is the data? HTTP? HTTPS? something proprietary? if it's HTTP can I cache any of the content? If it's HTTPS can i man-in-the-middle it to cache any of the content? How secure do you want it? Do you want traffic logging? complete traffic capture for non-repudiation? admissible in court? intrusion prevention? ...

Too many questions for a forum like this.

m.
 
You must have much more demanding feeds than I have, as ruynning IB/TWS during the day takes no more than 80mb.
I've got about 80 diff instruments up. And 3 DOMs. And a couple other live brokers. And other stuff....
Chap's

After you have sorted out your nuts & bolts, you need to find a provider for execution and leverage. Not retail leverage, but Prop. leverage, viz *100+
Nuts and bolts still worthy of a discussion. As far as leverage I would think/guess you would be trading futs rather than stocks. Margin rates are fine already for Asian markets. I mean you are not likely to get position sizes bigger than 100 contracts for most prop style intraday scalping in Asian markets. that's only $500,000 if a 100 lot trader hasn't that up their sleeves then there not likely to be a 100 lot trader :)
The simple answer is "it depends", there isn't enough detail here to propose anything,

Yeah fair enough was just a general question to feel my way out. will work out more and maybe PM at a latter date. Thanks
 
I mean you are not likely to get position sizes bigger than 100 contracts for most prop style intraday scalping in Asian markets. that's only $500,000 if a 100 lot trader hasn't that up their sleeves then there not likely to be a 100 lot trader :)

Or move to the ES, where they could get their 100 on the first level :eek:. Find a low-tax latin country or something and enjoy the scalability.
 
Yeah fair enough was just a general question to feel my way out. will work out more and maybe PM at a latter date. Thanks
No problem, would love to help someone out doing this kind of work. Would be a nice, reasonably challenging project to get up and going. Might also help put to use some of the spare bits & pieces of network kit I've picked up over the years.

m.
 
Did anything ever come of any of the ideas in this thread?
Other than the prop firms do any trading rooms similar to what amarcus is talking about in the first post exist anywhere in Australia?
 
Top