The absolute numbers below will change as technology progresses but the relative numbers should stay the same.

According to LMAX the relative cost of locks and CAS is as follows (May 2011):-

Clearly there are lots of issues with using float and double and using BigDecimal is not going to cut it an low latency financial world unless you are not really that low latency.  So what is the answer. As usual it depends, but here are some suggestions.

• Look after the pennies rather than the pounds. That is use cents rather than dollars, or what ever is the lowest precision in your world.
• Ensure intermediate calculations are as exact as possible so that rounding errors are not accumulated.
• Build or borrow a maths library to compare floating point numbers for instance (i.e float and double) with an epsilon value, i.e. a nearly equals method so that you can hide all the gory details in a library. Also add rounding and precision to the laundry list. Slightly surprised Java’s core libraries don’t cover this better.
• Use string representations on the wire rounded to appropriate levels of precision. Sure there is an overhead in conversion but at least its accurate.
• A reminder of sizes:-
  • Integer = 32 bits, 4 bytes = 2^31 =+- 2147483648 = 2.14748E+9
  • Long = 64 bits, 8 byes = 2^63 = +-9.22337E+18
  • Float = 32 bits, 4 bytes = +-3.5E+38 and +-1.4E-45, about 7 digits
  • Double = 64 bits, 8 bytes = +-1.7E+308 and +-4.9E-324, about 15 digits
  • String = words, 2 bytes, short

References

Java theory and practice: Where’s your point?
Java’s new math, Part 2: Floating-point numbers
Java float / double comparison

Here are some interesting low latency blogs and websites I’ve been reading:-

• http://psy-lob-saw.blogspot.co.uk Various Low latency
• http://mechanical-sympathy.blogspot.co.uk Various Low latency
• http://mentablog.soliveirajr.com Various Low latency
• http://jeremymanson.blogspot.co.uk Java Concurrency
• http://alexanderpodelko.com Performance Management
• http://bigocheatsheet.com Algorithm Speed
• http://highscalability.com Mainly for Web Site
• http://www.low-latency.com Aggregator website includes big data and analytics
• http://www.hftreview.com Aggregator website
• http://java-performance.com A bunch of tricks here
• http://vanillajava.blogspot.co.uk Writing-and-testing-high-frequency-trading-engines-in-java
• OpenHFT

Here are a list of web pages I’ve been reading recently:-

• Disruptor (lock free and garbage free) see also
• Pinning Threads
• Pinning Processes
• JIT Warming (but warming time can be complex)
• Memory Mapped IPC (within hosts)
• Lock Free Algorithms and Courses
• Inter Socket IPC with UDP (between hosts)
• SDP with Infiband (if you have the hardware)
• Java Large Pages (not always faster)
• Roll Your Own Everything including fix layer
• Immutable objects, but not for really low latency as need to create objects in the first place
• Cache friendly data structures cache misses are expensive
• O(1) or O(log2 n) at worst algorithms
• Single writer principle, all others are readers
• Number of threads = number of cores + 1
• Keep call stacks small and don’t use Spring as GC is harder with a big stack
• No locks or IO on the main flow
• Keep the flow simple and uncluttered
• Buy big machines with large RAM to avoid GC all together
• Use Unsafe for low level data manipulation and compare and swap.

In due course I will be adding posts for each of these topics.