Share
Embed
buildingSMART Technical Room Webinar IFC HDF5
Thomas Krijnen
May 26, 2020
Transcript
26 slides
-
IFC HDF5 Intro Use cases (large datasets, archival, point clouds,
sensor data / digital twin) Technicalities buildingSMART technical room webinar IFC HDF5 (binary) -
IFC SPF Advantages: human readable and machine agnostic Disadvantages: Slow
parse speed No partial reading no order on definitions no seeking ahead due to grammar buildingSMART technical room webinar IFC HDF5 (binary) -
In this presentation Level of Detail/Development http://bimforum.org/wp-content/uploads/2015/11/Files-1.zip Huge files with
heterogeneous contents (e.g point clouds, sensor data) Dramatically reduced read times buildingSMART technical room webinar IFC HDF5 (binary) -
Thomas Krijnen ✉ [email protected] @aothms Owner of AECgeeks, maintainer of
IfcOpenShell, researcher at TU Delft in the NLs and formerly TU/e. buildingSMART technical room webinar IFC HDF5 (binary) -
HDF5 Used in scientific computing EXPRESS mapping in ISO 10303
26 Adopted by NASA, Matlab, … buildingSMART technical room webinar IFC HDF5 (binary) -
Point cloud usage patterns acquisition and remodeling hybrid modeling as
built verification structural health monitoring Integrated separate storage buildingSMART technical room webinar IFC HDF5 (binary) -
Archival Long-term digital preservation properties Checksums Schema is encoded in
serialization (ifc4, ifc4add1, ifc4add2, ifc4add2tc1 => IFC4 schema identifier) Rapid partial reading and meta-data extraction buildingSMART technical room webinar IFC HDF5 (binary) -
Digital twins: real-time high frequency data in HDF5 packet table
Query and inference on semantic data + high resolution time series in standardized and portable container buildingSMART technical room webinar IFC HDF5 (binary) -
Layers of optimization Hierarchical storage of partial file chunks Partial
decompression (disk i/o typically bottle-neck) Binary, fixed-length storage buildingSMART technical room webinar IFC HDF5 (binary) -
Read more Thomas Krijnen, Jakob Beetz (2020) An efficient binary
storage format for IFC building models using HDF5 hierarchical data format, Automation in Construction 113, https://doi.org/10.1016/j.autcon.2020.103134 Thomas Krijnen, Jakob Beetz (2018) A SPARQL query engine for binary-formatted IFC building models, Automation in Construction 95, https://doi.org/10.1016/j.autcon.2018.07.014 Thomas Krijnen, Jakob Beetz (2017) An IFC schema extension and binary serialization format to efficiently integrate point cloud data into building models, Advanced Engineering Informatics 33 https://doi.org/10.1016/j.aei.2017.03.008 buildingSMART technical room webinar IFC HDF5 (binary)