Page History

One problem with 64-bit Java is the amount of Memory it occupies. As of 2011 we feel that all application profiles we have encountered do not require 64-bit Java as they often do not take advantage of using more than than 1-2GB of memory. Also, we have observed that when a Java Application has access to a large amount of memory, the application gets bogged down when garbage collection occurs.

As such, in many cases it is better to scale horizontally (running multiple JVMs) as opposed to using one giant JVM. This point is true for both 32-bit and 64-bit JVMs.

Just to be clear, you still should use a 64-bit Operating Systems. A 64-bit Operating System will allow a 32-bit JVM to come closer to the theoretical 4GB memory limit.

* Due to various additional constraints such as available swap, kernel address space usage, memory fragmentation, and VM overhead, in practice the limits will vary.

So I should never use 64-bit Java?

No, the point is to closely look at your application profile, user base and hardware and determine if you really do need to use a 64-bit JVM.

It has just been in our experience that often our profiles fall in,

• Application profiles use less than 4GB of memory and if they used more we use multiple JVM instances
• There is usually a 10%-15% hit in performance
• Large customer base
• Hardware with multiple CPUs handles multiple JVMs better

And let's not forget that if we want to switch to 64-bit it is simply a matter of recompiling.

Research

Overall there is a hit to performance and  when using a 64-bit JVM. Here are the top level reasons,

1. Native pointer in the system takes up 8 bytes instead of 4 (per Oracle)
2. Garbage Collection Pause (per Oracle)

According to, Software - Practice & Experience archive Volume 36 Issue 1, January 2006,

We conclude that the space an object takes in the heap in 64-bit mode is 39.3% larger on average than in 32-bit mode. We identify three reasons for this: (i) the larger pointer size, (ii) the increased header and (iii) the increased alignment. The minimally required heap size is 51.1% larger on average in 64-bit than in 32-bit mode. From our experimental setup using hardware performance monitors, we observe that 64-bit computing typically results in a significantly larger number of data cache misses at all levels of the memory hierarchy. In addition, we observe that when a sufficiently large heap is available, the IBM JDK 1.4.0 VM is 1.7% slower on average in 64-bit mode than in 32-bit mode.

Garbage collection may result in pauses in the application,

Remember, however, that this additional heap must be garbage collected at various points in your application's life span.   This additional garbage collection can cause large pauses in your Java application if you do not take this into consideration.