Hi, very interesting thread, some of you here seem very well informed on this topic and I wonder if I could ask for some advice
I live in the UK and trade in a variety of ways on markets on both sides of the Atlantic. Increasingly I am using algorithmic software for launching ultra short term positions which are created by software I have located on a New York server. For various reasons this cannot be relocated and the signals are generated faster by it being based in New York as this is where the raw data is located. This software currently fires the signal to my desktop application of this software which is on my 4 machines here in the UK and then launches a trade on my platforms here on my local machine in the UK also.
To be specific however with regards to one of my issues, one of my futures accounts is with MF Global based here in the UK which is hooked up to CQG Trader. So as things stand with this specific account my software in New York is firing me a signal which travels to the UK (54ms ping), initiates a trade on my platform, which then is routed via CQG and MF Global in London to Globex (probably another 150+ms at least). Obviously this is not the most efficient solution and only operates as such because of my other UK based and UK executed accounts.
What would your advice be to squeeze out any latency when using this New York based algorithmic software to trade CME products.
My initial thoughts are to get a VPS in Chicago to host the desktop version of the New York software and then to make this work I would also need my CQG platform on the VPS and broker be located in Chicago too.
I know it's a bit of a convoluted post but any thoughts would be very much appreciated!
I live in the UK and trade in a variety of ways on markets on both sides of the Atlantic. Increasingly I am using algorithmic software for launching ultra short term positions which are created by software I have located on a New York server. For various reasons this cannot be relocated and the signals are generated faster by it being based in New York as this is where the raw data is located. This software currently fires the signal to my desktop application of this software which is on my 4 machines here in the UK and then launches a trade on my platforms here on my local machine in the UK also.
To be specific however with regards to one of my issues, one of my futures accounts is with MF Global based here in the UK which is hooked up to CQG Trader. So as things stand with this specific account my software in New York is firing me a signal which travels to the UK (54ms ping), initiates a trade on my platform, which then is routed via CQG and MF Global in London to Globex (probably another 150+ms at least). Obviously this is not the most efficient solution and only operates as such because of my other UK based and UK executed accounts.
What would your advice be to squeeze out any latency when using this New York based algorithmic software to trade CME products.
My initial thoughts are to get a VPS in Chicago to host the desktop version of the New York software and then to make this work I would also need my CQG platform on the VPS and broker be located in Chicago too.
I know it's a bit of a convoluted post but any thoughts would be very much appreciated!

