We have tried to ensure that Instant JChem can handle large data sets with relatively modest memory usage. Instant JChem employs a sophisticated caching system to let it handle large data sets without all the data needing to be in memory at any one time.
In general this works very well, and you can for instance open the August 2000 NCI open database compounds, containing 250,251 structures, in SMILES format, and view and search this data quite effectively (try doing this in an Excel based package!).
However everything has its limits, and if you need to open very large data sets, or have lots of them open at the same time it is possible that Instant JChem may need more memory to operate effectively. Information on how to do this can be found in the Changing Memory Settings help topic.
As an approximate guide work on the basis that IJC needs 64MB for its own use and each million structures that you have will take up approximately another 100MB.
There is a memory toolbar that lets you monitor memory utilization. If this
is not displayed, right click in an empty area of the toolbar and check the
Memory checkbox.
![]()
This toolbar lets you view the current memory usage, and if you click on it it forces any unused memory to be freed up ('garbage collection'). Occasionally you may find that doing this may improve sluggish performance.
Using the memory monitor is a good way of detemining the maximum amount of memory to allocate to IJC:
Copyright © 1998-2009 ChemAxon Ltd. All rights reserved.