How latency is like a sore knee
As a running form instructor, I’ve seen many examples of how poor form leads directly to a sore or injured body. Poor alignment causes fatigue. Landing on your heels causes knee pain. And don’t get me started on how most people run up hills. Little inefficiencies don’t stop you in your tracks, but at the end of the day they can keep you from achieving your peak performance.
It’s a lot like having many little sources of latency in your trading system. Each weak link in the process keeps you from reaching optimal performance and profitability.
The best way to help people see what is wrong with their running form is to use video analysis and isolate in on alignment, landing points, hip rotation, arm swing, etc.
Finding and fixing the sources of latency in trading systems takes a keen eye and sophisticated tools, too. Today we announced a partnership with TS-Associates, a leading supplier of latency monitoring solutions. They’ve added support for Solace message routers in their TipOff product, which is itself a hardware appliance dedicated to non-intrusive latency monitoring. In what may be the logical opposite of slow-motion video analysis, their solution studies end-to-end traffic in your trading environment and gives you details on which components in your system represent opportunities to cut out latency.
We find there are three groups that are hot to trot for this kind of latency monitoring right now.
- High-frequency trading speed freaks. These are the guys that are challenging the laws of physics in an attempt to be first in line for information and liquidity, and building their trading strategy around that advantage. They have been getting lots of attention from the press and politicians lately, but the “speed for speed’s sake” crowd is actually quite small. It’s an inherently difficult way to succeed over time, because for any given strategy only one algo can be first and it’s easy for latency to be leapfrogged out of the blue.
- All the other algo traders. This group still cares about speed, but generally aren’t aiming to eradicate every last microsecond from their system. They have unique trading strategies that are differentiated by something more sustainable than speed. They just want to be competitive, and above all, demand low variance in latency (jitter). This group represents the bulk of the buyers of low latency solutions right now.
- Complex animals like investment banks and major exchanges. These are the firms with trading systems consisting of countless moving parts. They need latency monitoring to get their arms around their systems and prioritize what part to optimize or upgrade next. Many of their processes incur milliseconds worth of latency, not microseconds, but they still need to be competitive with peers to stay competitive.
For any of these groups, the most challenging part of latency monitoring is that the problems vary by scenario and traffic pattern. For example, in a given system a risk check may be the long pole in the tent during a single message lab test, while one of the algo engines may be the bottleneck at typical data rates, and the 1 GigE network links become the choke point during periods of peak traffic.
A product like TS-Associates’ TipOff gives you neatly summarized data so you can understand where to put your attention to support each of these conditions in a production trading environment.
If you’re a runner with sore muscles or joints, get yourself a video camera and some good advice. If you’re a financial institution concerned about latency, check out TipOff. It’s fascinating stuff.
Quants on wall street: Revenge of the nerds
I borrowed the headline above from the title of an interview at Yahoo! Finance today featuring Scott Patterson, the author of a new book called The Quants. His book tells the story of how mathematics took over trading on Wall Street.
I haven’t read the book yet (just downloaded on Kindle), but based on the teaser text at Amazon, I was disappointed to see that the author felt the need to tie his already interesting topic into the current financial malaise. The jacket text suggests he is blaming the quants for the 2008-2009 financial crisis. This will probably sell more books, but is an intellectually simplistic/sensationalistic/lazy argument in my opinion. I will reserve full judgment until I have read the book though.
Amazing how being anti-anything seems to be the preferred approach in books, newspapers and TV appearances these days.
Why power reduction goals are hard to meet
Every firm is concerned about power, but the problem continues to spiral out of control. Here’s a typical scenario that describe why things are the way they are:
The CIO makes what looks like the right declaration: “We WILL reduce power, no matter what”. Then good, honest projects come along that boost the top or bottom line and become exceptions to the “no matter what” part of the edict.
It is well documented that CIOs tend to last 2-3 years, and a typical IT worker will outlive several CIOs. Through time, these declarations take on the flavor of The Who’s immortal observation — meet the new boss, same as the old boss. Go on, click and waste the few minutes listening to vintage era Who. You know you want to, and it’s more energizing than drinking 2 cups of coffee. Not only that, but it immediately erases the memory of their Superbowl performance.
The best way we have seen to get the foot soldiers aligned with management objectives is to tie a meaningful piece of their compensation to it. Then everyone is aligned to find a way to make both kinds of “green” (more money and less power) happen. The good news is when half your annual pay is tied to anything, it becomes your focus. The bad news is this only works in financial services where bonuses are large enough to motivate. Having half of a government workers 10% bonus tied to power usage won’t change anyone’s behavior. They will become a bigger star, and get a bigger raise for delivering better or faster than they will for delivering cheaper.
Want a truly green datacenter? Turn off the servers.
You may hate the hype about “Green IT” as much as I do, but the problem it addresses is very real so don’t expect the term to go away anytime soon. There are some interesting stats on how out of control datacenter sprawl is getting in this article in eWeek Europe.
In 2020, data centres worldwide will consume about 450 billion kWh and their CO2 emissions (at about 330 million tones) will be equal to those of Portugal, Switzerland, Greece and Sweden combined. Their electricity bills will amount to nearly $45 billion (£26bn).
The demand for computing is always growing, so the only practical way to get more green is to do more work per watt.
This is a big reason why many customers are rolling out Solace technology. We generally get in the door on speeds and feeds, but we get picked and purchased based on our ability to reduce costs and shrink the datacenter. One Solace message router can replace the work being done by 10-30 servers running software-based messaging middleware, while using the power equivalent to about 1.5 servers, a power savings of 85-95%. That’s just power, remember, and doesn’t even reflect other cost-efficiencies like a smaller datacenter footprint and reductions in software licenses, software maintenance, linux maintenance contracts, and the server hardware itself.
It’s a win for everybody: faster applications for the operations teams and lower costs for the pencil pushers. Software-based middleware can’t demonstrate these results, and in fact is one of the key drivers of the “server sprawl” problem because scaling a software-based solution means splitting the work across more cores, more machines, etc.
The server vendors will forever push replacement cycles with a 10-15% more efficient machine, but it seems pretty clear that the most effective green strategy is finding ways to decommission the server racks altogether.
We’re bringing sexy (to the) back (office)
For the last few decades, from the early days of Gordon Gekko and Liar’s Poker to today’s focus on hedge funds and high frequency trading, the sex appeal has clearly centered around the front office. The press loves to write about it and we love to read about it. Every trade is like the proverbial iceberg, though — the trading decision is what everyone sees, but the majority of work happens in the murky waters below the surface where the considerably less sexy mid-and-back-office operations occur.
Last week Greg MacSweeney wrote a good article in Wall Street and Technology highlighting the back office as a new battleground for efficiency. After years of chasing zero latency for the front-office, the back office is comparatively archaic and badly in need of updating.
Read more