Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Trading Technologies / Patrick Rooney (Product Manager) - Ask Me Anything (AMA)
I'm looking for a new algorithmic trading platform and was wondering if I can ask you some technical questions regarding TT and X_Trader. My apologies upfront if the questions were already asked somewhere else in the forum (I swear I looked ).
I have my algo's already built in C++ and what I'm looking for is a way to connect/deploy them to a platform with the
lowest latency. I'm mostly interested to trade E-mini S&P 500 on CME.
1) I'm a bit confused about the difference between TT plaform and X_Trader. They both have C++ API, but, would that
be correct to say, one is meant for the full application development and the other for strategies and other plug-ins?
Which one do you consider the most suitable for high frequency trading?
2) I understood that if I developed my strategies in pure ADL, then I could somehow (magically) deploy them to a high-performance co-located environment without me having to connect and launch those strategies on the co-located server. Is the same possible with strategies that use C++ API?
3) Can you give me the most crude estimation or point me to some reports about the execution speed that I can expect from your co-located environment in CME for E-mini S&P 500? If my own algo was taking no more than a 1 micro-second to react to a tick (lets imagine), then what would be the latency of a round trip: tick occurring on the exchange, travelling to the co-located server, my algo reacting and issuing an order (1 micro-second), the order travelling to the exchange and getting placed? I know it depends on many factors but could you give me some numbers? (Lets also imagine that I'm using AMP Futures as an FCM)
4) Do you already support back-testing, in particular for strategies that use C++ API? I didn't find anything about it online, but if it's about to be released, are you going to support things like simulating execution delay, limit order fill rates and scenarios?
5) Does the data feed from TT, that I can access through API, provide Time and Sales for E-minis? Specifically, volume and sign of each market order with the corresponding time stamp? If so, what is the resolution of the time stamp, milliseconds, micro or nano-seconds?
Thank you very much in advance,
andrico
Can you help answer these questions from other members on NexusFi?
There's much to discuss here. I can give you some high level answers but you'll be better served speaking with our sales department for detailed answers.
TT Algo SDK has been available for many months and this is the API you would need if you wish to write and deploy your own algos on our co-located servers. https://www.tradingtechnologies.com/trading/apis/
As for backtesting, that's in development now and we'll have that in production later this year.
Yes, latency is incredibly difficult to speak about as there are a myriad of inputs that impact latency. You must speak to our sales team regarding your needs. https://www.tradingtechnologies.com/contact/
If you have any questions about the products or services provided, please send me a Private Message or use the futures.io " Ask Me Anything" thread
No worries, just wanted to ask. And while I'm at it: please consider showing delta (bid-ask market orders) on the VAP column. You already show colored volume bars, so I guess it wouldn't be hard to implement the delta. Thanks.
).
