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)
As zoltran points out you can not really improve latency (assuming that you haven't got actual faults anywhere) but you can often improve real world application performance, by tuning the transport layer to better accomodate the higher level process (client) to process (server) communications.
Tcp Acknowledge Frequency does make a difference in certain types of application (those that are also handling 'handshaking' at the application layer for example) especially if actual latency is already high. This is particularly true if the application is unaware exactly what the transport protocol is doing. It almost certainly will be. Most software engineers write to the appropriate windows API with little understanding of what is going on 'further down the stack'. Completely understandable thats the whole point of a layered protocol stack you don't have to worry about managing stuff at lower levels.
Worth mentioning ping is not a useful tool here as it operates at the network layer, so below this stuff. To test you would probably want to write some ninjascript to place and cancel an orders outside the market and record the time taken in a log file. Depends how far down the rabbit hole you want to go
Can you help answer these questions from other members on NexusFi?
That's kind of tough to answer. Like others here who already know, my experience came from working in the industry.( Mind you when I started out telegraph circuits where common and X.25 was state of the art. )That doesn't help you really. Wiki tells you a bit about TCP/IP and what various 'layers' do, all good background knowledge. it is worth understanding that framework as things make much more sense when you know what each layer does - it isn't going to help you directly though.
Also small p.s. on example of music players for computer.
I like to listen music at hi quality level (if not hi-end)
so, some 10 years ago, one friend of mine advised me one music player for computer and told : check it out, when I will install it and play you will think you've changed sound card, however I never used "bad or weak" computer at those times.
So, I did what he advised and...I was shocked difference between usual Media Player and that music player, even without it upgrade to its pro version.
Well, I used that for year and enjoyed it was like, say, Teac, but then a few months ago, don't remember what I've searched for, I found just amazing music player, when I played music with it I got impression I listen NAD which provides my favorite conception of sound.
The same with all other, if one is not searching for better or never did that or stopped do that, so the result will be.
Just noticed that their latest skin for the Web interface has a green 'light' next to the time. If you hover over this it tells you your latency with ZenFire.
This is useful as you cannot Ping ZenFire IP addresses.
I averaged 35ms to all of the listed hosts, which is pretty good considering it is just a DSL line. Where I live, DSL is the best I will have for quite some time, but it has been good enough.
what is this futures.io (formerly BMT) server ? Is it a hosting solution ?
I'm getting 33ms from my hosted server in Toronto and would love the snappier response time for my bots.
Pinging 64.202.118.5 with 32 bytes of data:
Reply from 64.202.118.5: bytes=32 time=33ms TTL=248
Reply from 64.202.118.5: bytes=32 time=32ms TTL=248
Reply from 64.202.118.5: bytes=32 time=32ms TTL=248
Reply from 64.202.118.5: bytes=32 time=32ms TTL=248
Ping statistics for 64.202.118.5:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss
Approximate round trip times in milli-seconds:
Minimum = 32ms, Maximum = 33ms, Average = 32ms
Mirus futures sells virtual machine hosting in 3 different packages ranging from $350-$550/month and claim a response time of less than 1ms since their servers are co-located with the exchange and beyond the DMZ.
I currently pay $60/month for a windows VM which does perfectly fine for trading 4-5 TF contracts (no slippage), but my goal is to be trading 20 contracts eventually and at that point I suspect a better response time might be more suitable for my 6 to 15 tick scalping bots.