Once an execution is started, as a caller I have no control over the amount of memory and CPU time that is being used. Do you think it would be reasonable to
- allow setting hard limits that will make additional allocations fail
- stop the execution after a certain number of instructions?
The execution limit could potentially be made possible without any such limit existing in the engine by exposing more internal api so instead of just invoking invoke and waiting for the result, some API for driving each step could be exposed.
Once an execution is started, as a caller I have no control over the amount of memory and CPU time that is being used. Do you think it would be reasonable to
The execution limit could potentially be made possible without any such limit existing in the engine by exposing more internal api so instead of just invoking
invokeand waiting for the result, some API for driving each step could be exposed.