Well said. In fact, my platform is Linux. My point is really that one probably shouldn't take a single factor and shoot an entire system down based on that single factor.
Progress Apama Algorithmic Trading
- Thread starter bidask
- Start date
Well said. In fact, my platform is Linux. My point is really that one probably shouldn't take a single factor and shoot an entire system down based on that single factor.
your firm's paying a rediculously high price for hypeware, thats the bottom line... obviously you guys can afford it therefore everyone's happy and well done to the apama girls... but if on 2nd thoughts you'd prefer this cash to stay in your bonus pool next year (pays for a few parking tickets doesn't it) you may want to look around more... thats all i'm saying, with 20+ years in finance etc...
[
...love it!
2cents, I appreciate your insights. I came to my current shop looking for something that would take the infrastructure off of my hands (having built the automated trading platform for the Portfolio Trading desk at my last shop), so that I could concentrate on my prop strategies. There, I had first-hand experience being pulled away from my prop attempts to add enhancements to the agency business, etc.
If you could show me a viable, reliable platform (combination of freeware tools) that could replace Apama, WHERE I WOULDN'T HAVE TO DO A BUNCH OF INFRASTRUCTURAL CODING, I'm all ears. My plan has always been to revisit things once I have been in full production for a while. It's another reason why I wanted to keep my prediction engine completely free from any vendor-specific strangle holds.
...love it!
2cents, I appreciate your insights. I came to my current shop looking for something that would take the infrastructure off of my hands (having built the automated trading platform for the Portfolio Trading desk at my last shop), so that I could concentrate on my prop strategies. There, I had first-hand experience being pulled away from my prop attempts to add enhancements to the agency business, etc. If you could show me a viable, reliable platform (combination of freeware tools) that could replace Apama, WHERE I WOULDN'T HAVE TO DO A BUNCH OF INFRASTRUCTURAL CODING, I'm all ears. My plan has always been to revisit things once I have been in full production for a while. It's another reason why I wanted to keep my prediction engine completely free from any vendor-specific strangle holds.
FredBloggs
Guest
dude. cum on!! if its $150k, thats hardly mega bucks. it may seem like big money if you do 2 lots in nq sat in your converted garage/home office, but for a professional outfit looking for professional returns to satisfy professional investors, then 150k is money well spent if it gives you the tools to get an edge.
btw, you dont get to that level by nickel and diming
cheers, and to bmferguson as well!
i know its not big money, pays for another grunt on a desk and headcount's hard to get though but thats another story...
its just that those costs add up, and for the wrong reasons...
give u an example, portware, at this bank am looking at right now, licence is $750K p.a., for some reason to the guy who signed a year or 2 ago, that made sense... today they've realized that the guts don't deliver as expected and are redeveloping all in-house (at a cost), therefore only keeping this turd for the GUI (to simplify)... thats a pretty expensive GUI at this rate... i'd say, if u want your algo / quant etc desk to make more $$$, start throwing away your portware, kbd, apama and other stupidly priced tools, stay away from the fidessa guys and whoever co-sponsors their "seminars"...
bmfergie, of course am not going to build u a solution, am busy enough building mine and we're v.likely not missing the same bricks anyway... but did u look at the cqg toolkit for instance? (thats not what i am using but i saw pretty good stuff there, from mates of mine, quants at another bank) and the hotbridge guys may have some other of the bricks u need for instance...
i know its not big money, pays for another grunt on a desk and headcount's hard to get though but thats another story...
its just that those costs add up, and for the wrong reasons...
give u an example, portware, at this bank am looking at right now, licence is $750K p.a., for some reason to the guy who signed a year or 2 ago, that made sense... today they've realized that the guts don't deliver as expected and are redeveloping all in-house (at a cost), therefore only keeping this turd for the GUI (to simplify)... thats a pretty expensive GUI at this rate... i'd say, if u want your algo / quant etc desk to make more $$$, start throwing away your portware, kbd, apama and other stupidly priced tools, stay away from the fidessa guys and whoever co-sponsors their "seminars"...
bmfergie, of course am not going to build u a solution, am busy enough building mine and we're v.likely not missing the same bricks anyway... but did u look at the cqg toolkit for instance? (thats not what i am using but i saw pretty good stuff there, from mates of mine, quants at another bank) and the hotbridge guys may have some other of the bricks u need for instance...
That's kind of the point. You're not going to build me a solution -- nobody is. You either pay for it by having someone deliver the bricks you're missing, or you hire someone to build them for you, or you build them yourself at a cost to you in terms of missed opportunity and potential burnout (as you know, the inertia involved in putting together a solid system is huge). No matter how you go about it, you pay. It's just how you decide to pay given your contraints. We're not all in the same position.
absolutely right
now one thing re burnout, unless say apama in this instance has got it 95%+ right in terms of your requirements (and that'd be a small miracle imo, but why not), i find the burnout in dealing with not-so-competent vendors / not-so-flexible platforms & toolkits is no less severe, when your focussed on delivering, ie results, not effort... my "solution" has always been to eliminate "clutter" fast, stay away from stuff thats loaded with too many bells & whistles (can't be good), and while i don't necessarily mind overpaying within reason, i want value... but yeah, in a corporate context, as long as you don't sink the boat, costs are less of an issue somehow... its only the shareholders' money after all
))))
can't resist sharing that one: the "bank" i am looking at, their IT spend for 2007... take a seat... $5bio...
can sure pay for a few apamas...
cheers
now one thing re burnout, unless say apama in this instance has got it 95%+ right in terms of your requirements (and that'd be a small miracle imo, but why not), i find the burnout in dealing with not-so-competent vendors / not-so-flexible platforms & toolkits is no less severe, when your focussed on delivering, ie results, not effort... my "solution" has always been to eliminate "clutter" fast, stay away from stuff thats loaded with too many bells & whistles (can't be good), and while i don't necessarily mind overpaying within reason, i want value... but yeah, in a corporate context, as long as you don't sink the boat, costs are less of an issue somehow... its only the shareholders' money after all
))))can't resist sharing that one: the "bank" i am looking at, their IT spend for 2007... take a seat... $5bio...
can sure pay for a few apamas...
cheers
Thanks for considering Esper. The company behind Esper is EsperTech Inc. providing support, services and training.