Currently more than 20 thousands of VO services are registered in the VO registry database. Keyword Search is the most popular way to find a resource. There can be a lot of ways to implement this capability, and the performance depends on how to index the document describing the resource metadata. Apache Solr is an open source search platform and uses the Lucene Java search library for full-text indexing. The document is indexed after the process that removes stop words (not adequate for a keyword e.g. a, the, and, is, are ... ) and stems a word to a root word (e.g. clustering to cluster). Thus for the query of a keyword cluster, the Solr search system will return documents that contain a word cluster, clustering, or clustered. Solr also can handle single- or multi-token synonyms and abbreviation. For an example, SN, supernovae, and SNe are translated to supernova in the process of indexing. So it increases the probability for a given keyword to hit a desired resource metadata. In order to incorporate this feature, we upgraded the registry service behind the JVO portal to the one based on Solr from XML based DB. The data from Nobeyama Legacy project were released from the JVO portal on 1st June in 2018. The data are now distributed also through the most recent VO standard interface called SIA-v2. The data of ALMA and Subaru telescope are also accessible through the SIA-v2. We are now working on distributing the data of Hyper Suprime-Cam SSP/DR1 and also the data of Hitomi satellite under the collaboration with C-SODA/JAXA. We present those our recent development on Japanese Virtual Observatory system.
Link to PDF (may not be available yet): P13-17.pdf