Energy monitoring of an allocation

[Jacob Goldverg]

Hi Everyone,

I have a few questions which are all interrelated regarding experimentation for energy monitoring with WAN’s.

1. Is there any way to monitor the energy consumption of a VM allocation via Intel RAPL? Usually I find the latest readings in /sys/class where there are rapl specific files that report the # of joules. Is this available in the VM as typically in cloud providers like AWS this is not provided??

2. Currently my WAN experiments utilize traceroute to get a list of IP’s where I then geolocate the IP’s. In Fabric in the IP addresses are all private hence I was wondering if there some other API that I could utilize to get the coordinates of roughly where the hops are located.

3. Coming back to hops, is there anyway I could capture the energy consumption of a hop? Today I currently utilize Simgrid where I have predefined the byte to joule ratio that lets me estimate the energy used by that specific hop. I could use this same approach with Fabric assuming I have the make and model of routers & switches used in my network path, but it would more ideal if I could query some API to let me know the real time energy consumption of my data flow.

 

Thank you for answering these loaded questions.

 

Best,

Jacob

[Komal Thareja]

Hi Jacob,

At the moment, energy consumption measurements are not passed into the VMs. I’ll bring this up in our planning meeting so it can be considered for inclusion in a future release.

For location information, we currently expose the geo-coordinates for all FABRIC sites, which you can retrieve using fablib.list_sites(). One possible approach to determine the location of hops is to map IPs → Sites → Locations.

Could you share your slice ID or specify the type of network service you’re using for your WAN experiment?

Best,

Komal

[Jacob Goldverg]

Hi Komal,

Right now I do not have an allocation on fabric as I am simply exploring how I can create an allocation that would have such details. For my work its essential that I get the min/avg/max power utilization from RAPL(or something similar) for routers and for the VM itself. Is there anyway I can query the hardware type to get such numbers at least as a baseline?

 

My advisors name is Tevfik Kosar, and I believe back in April 2025 he got a renewal of our allocation but we do not see it updated in the UI. Would you be able to confirm that it went through? The project ID is 8c5ce2ce-cfc6-4df6-8141-bd4de2bdece5

 

Best,

Jacob

[Komal Thareja]

Hi Jacob,

Thank you for sharing the details. I discussed the energy consumption measurement topic with the team earlier today.

As mentioned before, we do not currently support energy consumption measurements on the VMs. The network team also confirmed that such measurement capabilities are not available on the network devices.

For identifying the geolocation of hops, we recommend using the L2PTP (Layer 2 Point-to-Point) network, where the user explicitly defines the network path by specifying the SITEs for each hop. You can then use fablib.list_sites() to obtain the geo-coordinates.

Please refer to this artifact for guidance on setting up an L2PTP slice:
https://artifacts.fabric-testbed.net/artifacts/7e439627-96be-45e0-ab67-50bb607f06e4

 

Also, regarding the renewal request, I see that it was submitted yesterday. Michael will follow up on that through the ticket.

Best,

Komal

• This reply was modified 2 weeks, 4 days ago by Komal Thareja.

[Author]

Posts