System-on-chip Test Scheduling Research Articles

Window-based peak power model and Particle Swarm Optimization guided 3-dimensional bin packing for SoC test scheduling

This paper addresses the issue of power-aware test scheduling of cores in a System-on-Chip (SoC). While the existing approaches either use a fixed power value for the entire test session of a core or cycle-accurate power values, the proposed work divides the power profiles of cores into fixed-sized windows. This approach reduces the number of power values to be handled by the test scheduling algorithms while reducing the amount of pessimistic over-estimations of instantaneous power consumption. As a result, the power model can be integrated with more exhaustive meta-search techniques for generating power constrained test schedules. In this paper, the proposed power model has been integrated with a Particle Swarm Optimization (PSO) based 3-dimensional (3-D) bin packing technique to generate test schedules. Experimental results prove the quality of the approach to be high compared to the existing scheduling techniques.

A Graph-Based Approach to Power-Constrained SOC Test Scheduling

The test scheduling problem is one of the major issues in the test integration of system-on-chip (SOC), and a test schedule is usually influenced by the test access mechanism (TAM). In this paper we propose a graph-based approach to power-constrained test scheduling, with TAM assignment and test conflicts also considered. By mapping a test schedule to a subgraph of the test compatibility graph, an interval graph recognition method can be used to determine the order of the core tests. We then present a heuristic algorithm that can effectively assign TAM wires to the cores, given the test order. With the help of the tabu search method and the test compatibility graph, the proposed algorithm allows rapid exploration of the solution space. Experimental results for the ITC02 benchmarks show that short test length is achieved within reasonable computation time.

Formulating SoC test scheduling as a network transportation problem

A formulation of core-based system-on-chip (SoC) test scheduling as a network transportation problem is presented. Given a set of tests, with demands for transportation of test bits (either for test stimuli or test response) and unrelated parallel test resources (e.g., test access mechanisms or built-in self-test engines), the authors determine the start times and resource mappings of all the tests such that the finish time for the complete SoC test is minimized. The problem is NP-hard and they present an approximation algorithm using a result from the solution of the single source unsplittable flow problem. The proposed method uses the number of test bits that need to be transported for a test as the invariant and is hence relatively independent of the test application and execution model. Experimental results on benchmark SoCs demonstrate that their method outperforms the state-of-the-art integer linear programming formulations, not only in terms of schedule quality, but also significantly reduces the computation time.