SeqMonk works with very large data sets (both genome annotations and mapping data), and it therefore requires a large amount of RAM to run.
Whilst most aspects of the program can be altered whilst SeqMonk is running, the total amount of memory available to the program is determined when it is started. The SeqMonk launcher program will try to determine the optimal amount of memory to assign to SeqMonk when it starts, but you may need to take steps to adjust this in some cases.
The default configuration of SeqMonk is to assign the program 2/3 of the physical memory installed on your machine. Thus on a machine with 6GB of RAM, 4GB will be allocated to SeqMonk. Because of overheads in java itself the amount of available memory reported in the welcome screen will be around 80% of the originally allocated memory.
This default allocation is then limited in some cases. If you are running a 32 bit operating system, or a 64 bit operating system but a 32 bit version of java then the maximum memory allocation is limited to 1.5GB however much RAM is installed. If you want to make use of more memory than this you will need a 64 bit OS and version of java. On 64 bit machines running 64 bit java the top memory limit for SeqMonk is 10GB, and under most circumstances it's unlikely that you'll benefit from using more than that.
If you do want to manually set the SeqMonk memory allocation you can do this. What you can adjust is the upper limit of the memory allocation which the launcher will allow. You'll still be restricted by the rule of not allowing more than 2/3 of the physical memory in your machine.
You can alter the upper memory limit from Edit > Preferences > Memory. You will see an "inital memory setting". Setting this to zero will automatically assign memory and is strongly recommended. You can however assign your own value to increase the maximum assigned memory. The value in this field is in MB so if you wanted to assign 20GB you'd need to put (20 x 1024) = 20480 in the box.
This setting will not take immediate effect but will only do something the next time you launch seqmonk.
The other option for changing memory limits is to force a memory value on the command line. This value is not sanity checked or limited so if you set it too high then seqmonk will completely fail to launch. It is not subject to the constraint of being only 2/3 of your physical RAM though.
Changing the memory allocation is as simple as passing an extra parameter when launching the program. Instead of starting seqmonk by just running 'seqmonk' you would run 'seqmonk -m XXXX' where XXXX is the amount of memory (in MB) you want to use.
On OSX to be able to access this option you'll need to start seqmonk from the terminal application, and to use the launch script directly. If SeqMonk is installed in your applications folder then you'd need to do something like: cd /Applications/SeqMonk.app/Contents/MacOS ./seqmonk -m 20480