This paper proposes a new approach to parallel applications development, 
showing high speed and low labour intensiveness of creating new parallel 
applications and parallelizing existing ones. This approach introduces 
new semantics of procedures and their parameters, allowing user to declare 
potential parallelism in a natural and simple way and parallel computing 
framework to use such information in efficient way. The basic principles 
of the proposed approach and possible ways of their implementation are described. 
The problem of analysis of time characteristics of execution of parallel 
programs in this technology is also described. An approach that combines 
profiling and asymptotic execution time analysis is being proposed. 
It can be used for estimating program execution time, finding and eliminating 
bottlenecks, estimating the power of cluster needed to execute 
the program within given time limit.