The three-dimensional (3-D) technology offers a new solution to the increasing density of integrated circuits (ICs). In this work, we propose novel scan architectures for 3-D IC pre-bond and post-bond testing by considering the interconnection overhead of through-silicon-vias (TSVs). Since hotspots in 3-D ICs often cause performance and reliability issues, we also develop different test ordering schemes for prebond and postbond testing to avoid applying test vectors that could worsen the temperature distribution. Experimental results show that the peak temperature can be lowered by 20% with the 3-D scan tree architecture. When combined with the test ordering scheme, the 3-D scan tree can further reduce peak temperature by over 30%.
ACM Reference Format:Dong Xiang and Kele Shen. 2014. A thermal-driven test application scheme for pre-bond and post-bond scan testing of three-dimensional ICs.
3D technology for networks-on-chip (NOCs) becomes attractive. It is important to present an effective scheme for 3D stacked NOC router and interconnect testing. A new approach to testing of NOC routers is proposed by classifying the routers. Routers with the same number of input/output ports fall into the same class. Routers of the same class are identical if their tests are the same. A test packet is delivered to all the identical routers by a simple unicast-based multicast scheme. It is found that the depth of the consumption buffer at each router has great impact on the test delivery time because test application and test delivery for router testing cannot be handled concurrently. Test delivery must set a router to operational mode. A mathematical model is presented to evaluate the impact of consumption buffer depth on the test delivery time. A new and simple test application scheme is proposed for interconnect testing. Some interesting extensions are presented for further test time reduction and thermal considerations. Sufficient experimental results are presented by comparison with one previous method. The proposed method works for single stuck-at, transition, even small delay faults at routers, and single bridging faults at physical, consumption and injection channels.
The technology of three-dimensional (3D) SoCs is emerging as a promising approach for extending Moore's Law. Managing test architecture design and optimization of 3D integration are crucial challenges. In this paper, we propose a reconfigured test architecture optimization for 3D SoCs, including a novel scheme to minimize the prebond test time and Known-Good Stack (KGS) test to guarantee the yield of 3D SoCs. Experimental results on ITC'02 SoC benchmark circuits show that our scheme reduces the total test time by around 23% on average and nearly 30% in maximum compared with one baseline solution.
With the growing complexity of embedded VLSI products, traditional System-on-Chip (SoC) are facing severe challenges in the aspects of communicating speed and scalability. Network-on-Chip (NoC) has emerged as a viable alternative. In NoC design, application mapping is one of the most holistic researching dimensions, which maps the cores in the application to the routers in the NoC platform. Application mapping problem usually aims to reduce communication cost and power consumption of the overall system. In this paper, we focus on application mapping onto mesh network, and propose a novel two-phase heuristic algorithm. The first phase attempts to explore the potential searching spaces, while the second phase focuses on exploiting the local optima within the searching basin. To verify the effectiveness of the algorithm, this paper performs a quantitative comparisons between our proposed method and the existing mapping methods under both real application and custom generated application benchmarks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.