Tibco

Quote from jjw:

The Macy's analogy is good but i think that it is not quite the point beacuse it is based upon the folloing premises:
1. all things being equal; and
2. that the tailor on saville row is always as good or better as any tailor in the world.

i address premise 2 first. the middleware/messaging produced by my firm has been used by firms like GS and Getco after having been tested against their own home grown messaging/middleware. the bake-off was not a middelware/messaging comparison but a comparison of a trade particular to the trading firm - our rimplementaiton .vs theirs. we won, they chose us over their own. this is not to say that we are better (we were in that instance) but just because GS is GS or Getco is Getco, does not mean that they always have the best. they certainly have money to get the best but it could be that the best resides outside their respective firms.

with respect to premise 1 - all things being equal - all things are never equal, it is just a matter of the latency tolerance of a trading algorithm to keep it in the game or not(assuming it is otherwise a profitable algorithm). consider that some feeds (equity options in particular) distribute their market data via tcp. in such a case someone gets the data first - all others are not first. depending upon the number of recipients of the market data, latency realized by the last recipient can be several milliseconds. unfortuantely such an exchange will not disclose to its recipients where they are in the market data distribution list. now consider that most market data these days is distributed via udp multicast. a common myth is that every one on the same switch gets the data at the same time. each recipient may get the data much closer in time, say 10 microseonds difference, but again, someone is first and some one is last. in the world of ultra low latency trading, 10 microseconds makes a difference.

as this kind of latency is beyond a recipient's control (and also may be different from day to day) a trader, even a GS or Getco, in my opinion, is still better off focusing on what they can control and on what can save them milliseconds and leave the microseconds to the microsecond maniacs.
More...

Good post.

This was back when I was at LEH...

The JMIS (Japanese Market Interface System) quote's latency started increasing for about a week. And the clients and some of us trading in-house complained to the Connectivity team. Fortunately, the team had an in-person meeting with the TSE so they forwarded the complaint to Tokyo Stock Exchange.

Few days later the quotes were fresh and clean...

All things being equal??? No they're not, all you have to do is complain. Though, the difference is how to make them listen to your complaints.

Shit happens.
 
I believe a technology person should pick the "right tool for the right job". when I was starting this trading firm, I was just poor, can't afford even to spend 20k on software, after writing some astronomical checks to buy Exchange Memberships.

So yes, it is strangely satisfying to see my fills show up on the exchange market data update first. To do that, we control *every* facet of the trading platform. Just today, forget microseconds, I was impressed with a new developer (a young kid), shaved *23 instructions* from a well used function (very creative use of bitwise rotation). that's 0.01 microsecond to you sir, on my 2 ghz thinkpad, when we all sat down to look at the profiler output.

However, when I was a CTO, I had this running joke of "2M" standard answer for any initiative. It is true! Because take the resources to evaluate the technology, do a prototype, get the necessary hardware in, get the signoffs, build a business case, etc. It may seem inefficient and bureaucratic, but these process controls are necessary in a large firm.

So I don't know if there is necessarily a "right answer", it is all just a matter of emphasis. Some firms like to "off load" the "system maintenance nightmare" by buying most software off the shelf. Some firms like to be similar to a research center, where enormous resources are spent on looking at new technologies, etc, often with no result to show for.
 
I used to work FOR Tibco, my good friend is still an architect there and the company's bread and butter is no longer the RV middle ware but B2B integration like Business connect product line. We used, way back to hack together trading platforms using RV and they were almost always disasters and (not only because the lagg), back then we were the only game in town and Reuters as big daddy. Now it is all different......
I however concur with the poster(rufus?) in reference of "build your own"
as an insane idea for corporate solution for big money trading......
since then I used other middleware products (names won't mention and they suck worse then TIBX).
As for the lag it is not always an issue, the last gig I was with a trading desk now defunct (Robertson Stevens), where sales traders would take the orders and route to the floor where VWAP them for a nice average price...now where in here a 2 sec delay will ever hurt you?

P.S. TIB/RV was NEVER meant for ET type "automated trading", EVER that is just a silly notion......
 
Quote from loza:


As for the lag it is not always an issue, the last gig I was with a trading desk now defunct (Robertson Stevens), where sales traders would take the orders and route to the floor where VWAP them for a nice average price...now where in here a 2 sec delay will ever hurt you?

P.S. TIB/RV was NEVER meant for ET type "automated trading", EVER that is just a silly notion......
More...

We're talking about High Frequency Trading, not Worked Orders....
 
...and I am saying that clearly TIB/RV is not (and NEVER BEEN) for that but then MOST BIG MONEY MIDDLE WARE market is not for daytrading/scalping/picking up nickels before the bulldozer ET type BS either......TIB is doing just fine WITHOUT fishing for the auto/daytrade biz! just look at their stats and earnings.......
Market Cap: 1.17B
P/E (ttm): 22.76
EPS (ttm): 0.29
Quote from TSGannGalt:

We're talking about High Frequency Trading, not Worked Orders....
More...
 
I looked into 29West to see what the costs are. $70k a year! Since I started this thread, I have combed the web for other possibilities. I am studying the Ace threads library as middleware:

http://www.cs.wustl.edu/~schmidt/ACE-overview.html

It appears to be as close to programming at the OS level as possible without giving up too much control and performance. The only "problem" is that it is strictly C++, and there is a java wrapper. No .Net. To those that don't believe that .Net loses any performance and need to stay within .Net, this won't work, at least not in a mixed environment.

If any of you guys have experience with it I would be interested in hearing it.
 
Quote from tetramorium:

Do you know of anyone having implemented a production system based on Mina?

Thanks
Tetramorium.
More...

QuickFix/J is built on Apache Mina and that is certainly used in production.

As an alternative to Mina, you might want to have a look at JBoss NETTY: http://www.jboss.org/netty/

Suggest you Google the history of these projects how they are related and performance comparisons. Both are open source so you can examine the code too.
 
The hardcore part of the microsecond maniacs have implemented their algos in FPGA. Very useful for allowing highly parallel processing so every quoted instrument can have it's own hardware to implement the quoting /trading algo. Messaging/middleware as far as it exists is just a bus on the same piece of silicon.

Taking that into account, a discussion of which software to use is really one step behind the curve. Perhaps we will see a thread on which FGPA to use as time goes on. Personally, I think it all starts to get so expensive that it is probably cheaper to gain an informational edge rather than trading on a latency advantage alone.
 
You must log in or register to reply here.
Back
Top