LHC@home

Every VM using Condor connects to vccondor01.cern.ch:9618 once per minute.
With thousands of concurrently running VMs out in the field this produces a high connection rate on the condor server.
As a missing condor ping at startup leads to a VM shutdown I wonder if the period between the regular contacts can be extended.

It should scale but is something we could look at reducing. The test should now try three times before failing. Let's see if that helps.

It should scale but is something we could look at reducing. The test should now try three times before failing. Let's see if that helps.

Must have been a very important question to get 2 answers.
;-D


Much more important:
Congrats, Laurence.
Your first masterpiece?

It should scale but is something we could look at reducing. The test should now try three times before failing. Let's see if that helps.

If there is any way to reduce traffic, such as storing redundancies locally, streamlining the size of each packet, using more compression, etc it would be appreciated.

Currently I have a typical ~300kb bandwidth remaining of a 25,000kb ISP internet plan as the Theory WU's crunch.

The next step up data plan would more than double my monthly ISP service fees.

If you have a small cluster at home, it should be possible to setup a local squid cache to reduce the external traffic. A few people have looked into this but we don't have any detailed instructions for anyone to follow. I will create a thread in Number Crunching about this.

If you have a small cluster at home, it should be possible to setup a local squid cache to reduce the external traffic. A few people have looked into this but we don't have any detailed instructions for anyone to follow. I will create a thread in Number Crunching about this.

Thankyou.

computezrmle reminded me of benefits of using a proxy server and some strategies of setting one up.