Search Mailing List Archives
[bioontology-support] BioPortal SPARQL endpoint locally installed?
manuelso at stanford.edu
Wed Feb 27 11:19:42 PST 2013
On Feb 27, 2013, at 12:59 AM, Erick Antezana <erick.antezana at gmail.com> wrote:
> Hi Manuel,
> many thanks for the information and pointers. I will get this
> installed locally.
> On more request: could you please provide some details about the
> hardware you have? I would like to have an idea on the "machine's
> size" (that is, memory, hard disk, CPU, OS, ...) so that we could
> better enable such a local instance?
The hardware we have for the open SPARQL public endpoint is a 4+1 cluster (4 backends + 1 master). Two CPUs per backend and 4 for the master (spec: Intel(R) Xeon(R) CPU @ 2.93GHz). RAM: master --> 16G , backend --> 4G. OS: redhat. Hard disk … we use some network NFS system with RPM disks.
The thing is that this infrastructure is built on top of VMWARE server so in a way it is shared with other apps.
The hardware you'll need really depends on the query workloads that you are expecting to have. In the 4store google-group there are some discussions that talk about this topic.
> On 22 February 2013 18:04, manuelso <manuelso at stanford.edu> wrote:
>> Hi Erick,
>> You can use any triple store. We use a fork of 4store that supports rdfs:subPropertyOf reasoning, this is available here:
>> If you want to install the web fronted, this is the project:
>> To populate the database you can download the RDF files using the REST RDF download service:
>> On Feb 22, 2013, at 2:57 AM, Erick Antezana <erick.antezana at gmail.com> wrote:
>>> is there a way to get a locally installed instance of the BioPortal
>>> SPARQL endpoint?
>>> Which triple store is used?
>>> bioontology-support mailing list
>>> bioontology-support at lists.stanford.edu
More information about the bioontology-support