CCL Home Software Community Operations |
About HTCondor at Notre DameHTCondor is a distributed batch computing system used at many institutions around the world. HTCondor allows many users to share their computing resources, thus giving everyone access to more computing power than any one person could afford to buy. In particular, HTCondor is good at harnessing idle cycles from under-used machines such as desktop workstations. HTCondor is different from other batch systems that you may have used, because it protects the interests of everyone involved. If you own machines in the HTCondor pool, then you get to decide who uses your machines and when they get used. You can even kick running jobs off of your machines or temporarily disable HTCondor if you like. It's your machine, after all! Conversely, users that submit jobs to HTCondor can specify their own requirements on what machines to use. You can require the use of one particular machine, or any from a particular cluster, or any with a certain amount of memory or processor type. Of course, if you are borrowing machines owned by other people, then you must accept the possibility that some jobs may be kicked off and must run elsewhere. So, the best way to use HTCondor is to submit lots of jobs and take advantage of all the idle cycles, wherever they may be. If you are doing simulation work, don't submit one simulation, submit one hundred or one thousand variations at once. HTCondor cannot guarantee that any single job will finish quickly, but it will allow you to accomplish more overall work than if you used just your own machines. Thus, HTCondor is called a high throughput computing system.
Related Links |