Next: 1.5 Current Limitations
Up: 1. Overview
Previous: 1.3 What is Condor
- Checkpoint and Migration.
- Users of Condor may be assured that
their jobs will eventually complete even in an opportunistic computing
environment. If a user submits a job to Condor which runs on somebody
else's workstation, but the job is not finished when the workstation
owner returns, the job can be checkpointed and restarted as soon as
possible on another machine and will keep on executing in this manner
until the job is completed. Condor's Periodic Checkpoint feature can
periodically checkpoint the job even in lieu of migration in order to
safeguard the accumulated computation time on job from being lost in the
event of a system failure (such as the machine being shutdown, a crash,
etc).
- Remote System Calls.
- In Condor's Standard Universe execution
mode, the local execution environment is preserved for remotely
executing processes via Remote System Calls. Users do not have to worry
about making data files available to remote workstations or even
obtaining a login account on remote workstations before Condor executes
their programs there. The program behaves under Condor as if it was
running as the user whom submitted the job on the workstation where it
was originally submitted, no matter on which machine it really ends up
executing on.
- No Changes Necessary to User's Source Code.
- No special programming
is required to use Condor. Condor is able to run normal UNIX programs.
The checkpoint and migration of programs by Condor is transparent and
automatic, as is the use of Remote System Calls. These facilities are
provided by Condor
and, if these facilities are desired, only requires the user to re-link
their program, not recompile it or change any code.
- Sensitive to the desires of workstation owners.
- ``Owners'' of
workstations have by default complete priority over their own machines.
Workstation owners are generally happy to let somebody else compute on
their machines while they are out, but they want their machines back
promptly upon returning, and they don't want to have to take special
action to regain control. Condor handles this automatically.
- ClassAds.
- The ClassAd mechanism in Condor provides an extremely
flexible and semantic-free, expressive framework for match-making
Resource Requests and Resource Offers. One result is that users can
easily request practically any resource, both in terms of what their job
requires and/or what they desire for their job if available. For
instance, a User can require their job run a machine with 64 megs of
RAM, but state a preference for 128 megs if available. Likewise, machines
could state for example in a Resource Offer ad that they prefer to run jobs
from a certain set of users, and require that there be no interactive user
activity detectable between 9 am and 5 pm before starting a job. Job
requirements/preferences and resource availability constraints can be
described in terms of powerful, arbitrary expressions, resulting in
Condor being flexible enough to adapt to nearly any desired policy.
Next: 1.5 Current Limitations
Up: 1. Overview
Previous: 1.3 What is Condor
condor-admin@cs.wisc.edu