Direction-Optimizing Breadth-First Search with External Memory Storage

Authors: Shuli Hu, Nathan R. Sturtevant

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental To evaluate the effectiveness of DEBFS we experimented with Rubik s Cube, Chinese Checkers, Top Spin, and the Pancake puzzle. We ran our experiments on a 16-processor 2.4GHz Intel Xeon E5 server with 128GB of RAM, two 8TB disk drives configured as a RAID drive, and one 1.5 TB SSD.
Researcher Affiliation Academia 1Northeast Normal University, Changchun, Jilin, China 2University of Alberta, Edmonton, AB, Canada
Pseudocode Yes Pseudo-code for a parallel version of DEBFS is shown in the Algorithm 1.
Open Source Code No The paper does not provide any explicit statements about releasing source code for the described methodology, nor does it include links to a code repository.
Open Datasets No The paper conducts experiments on well-known problem domains (Rubik's Cube, Chinese Checkers, etc.) for which Pattern Databases are built, but it does not provide concrete access information (links, citations with authors/year) for publicly available datasets used as input to build these PDBs, nor does it state that the PDBs themselves are publicly released.
Dataset Splits No The paper does not provide specific details regarding training, validation, or test dataset splits; it focuses on building Pattern Databases for known problem domains.
Hardware Specification Yes We ran our experiments on a 16-processor 2.4GHz Intel Xeon E5 server with 128GB of RAM, two 8TB disk drives configured as a RAID drive, and one 1.5 TB SSD.
Software Dependencies No The paper mentions software components and parallelization, but it does not provide specific version numbers for any libraries, frameworks, or compilers used in the experiments.
Experiment Setup Yes We vary the memory and number of buckets used in each experiment, to simulate performance on machines with less RAM. The experiments in this paper use four bits per state... Table 1 provides details such as 'Buckets' and 'RAM used' for various configurations.