-
Notifications
You must be signed in to change notification settings - Fork 0
HowTo
If you only need a command-line access (no GUI) then you may SSH to csl2.cs.technion.ac.il and from there SSH to one of our TDK machines (note that t2 cannot be accessed from outside the Technion).
Alternatively, and in particular if you need a GUI access to the machine, you need to get a VDI (Virtual Desktop Infrastructure). Follow the guidelines explained in the system site.
You may of course use the VDI as described above. In case you only need SSH access, then login to csm.cs.technion.ac.il, and from csm you may SSH to the target machine.
There are plenty of SSH clients available for Windows. We strongly recommend MobaXterm. This is a very convenient client, that also allows to open Linux graphical apps in your Windows environment.
A new infrastructure that we use called VRA, allows to create a VM independently via a dedicated Web interface. To create a VRA VM you need to:
- Send a mail to Oren, asking for a permission to create VMs.
- When approved, follow the instructions explained in the System's site. Note that steps 1-2 are not relevant.
If you need to execute a long-running process in your Linux machine, use the screen tool. Here is a short video tutorial explaining the basics. You may also check this written tutorial.
If you need to execute SPARQL queries against the DBPedia or JBS datasets, we recommend on using our eLinda endpoint that is available at http://tdk-p6.cs.technion.ac.il:8081/search.html.
Settings:
- Set the graph name (second setting on the right). Use http://dbpedia.org for DBPedia, and http://jbs.technion.ac.il for JBS.
- For your convenience, you may define prefixes using the Prefixes button. Recommended prefixes for JBS are:
- PREFIX jbr: http://jbs.technion.ac.il/resource/
- PREFIX jbo: http://jbs.technion.ac.il/ontology/
Note: this google document contains popular queries for the JBS dataset.
If you use JavaScript then you may want to utilize the API written by Tal Yahav as explained here.
If you wish to access the endpoint from Java code (or from Android), then you should probably use the Jena Query API.