Search Mailing List Archives
[bioontology-support] NCBO memory problem
jean.paul.ik at gmail.com
Tue Dec 6 11:21:19 PST 2011
Thanks again, I downloaded the 0.4 ncbo appliance, and this problem is not
2011/12/6 Jean-Paul <jean.paul.ik at gmail.com>
> Thank you.
> I am giving only 2G of memory to the VM. But I see that the Xmx parameter
> of tomcat (/etc/init.d/tomcat) is set to 4500m. So is this a minimum
> requirement for the app to work properly?. I only have one user accessing,
> and very few ontologies. (for instance this one:
> http://www.w3.org/2005/Incubator/ssn/ssnx/ssn.owl). I load it, parse it
> and it works well. But when i browse the terms in the tree, the application
> dies after navigating some of the tree elements.
> 2011/12/5 Paul R Alexander <palexander at stanford.edu>
>> Memory usage will vary on how you are utilizing the Virtual Appliance and
>> exactly what is happening on the system at any given point. The portions of
>> the application that tend to use a lot of memory include: Lucene search
>> index, ontology parsing, and the Rails application when experiencing
>> mid-high traffic loads (~20 concurrent users).
>> We haven't heard of other users having memory issues, but again it is
>> very dependent on how you are using the Appliance. Your virtualization
>> environment should include options for increasing the memory available to
>> the Appliance, so starting there is what I would do.
>> Paul R Alexander
>> Web / UI Developer
>> NCBO BioPortal
>> Stanford Center for Biomedical Informatics Research
>> On 12/2/11 12:03 AM, jpcik wrote:
>>> I am using v 0.3 of NCBO. We followed the instructions and the
>>> a couple of weeks ago. It seemed tu run fine, but I noticed that
>>> Bioportal becomes unresponsive after a while, (we get database
>>> errors constanly). I am under the impression that it might be a memory
>>> problem ( the VM is running on a machine with only 3GB of memory). Does
>>> Bioportal require more than 4G of memory to run properly or maybe the
>>> of our problem is something else. I thought about that because you have
>>> this heavy cache implementation that might require lots of resources....
>>> View this message in context: http://ncbo-support.2288202.**
>>> Sent from the NCBO Support mailing list archive at Nabble.com.
>>> bioontology-support mailing list
>>> bioontology-support at lists.**stanford.edu<bioontology-support at lists.stanford.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the bioontology-support