We implemented solution based on Sphinx Search to replace default search for a large B2B portal. The portal was based on PHP and was using mysql fulltext for search the available products and other content. There were more than 30,000,000 products for search. This was causing a lot of load on database servers as the fulltext index on tables was growing rapidly. We designed and implemented new Sphinx indices for each content type and started using these instead of mysql. The positive impact was visible immediately as performance and quality of the search was greatly improved. Load on database servers was reduced as well. We were eventually able to reduce number of database servers which where being used for the fulltext searching, this made positive impact on cost of servers. Conversions from search pages improved which automatically helped in increasing revenue.


Waseem is consultant for Elastic Stack. He is Elastic Certified Engineer. Has years of experience with Elasticsearch, Solr, Wazuh, Sphinx Search, Manticore Search, OpenSearch and full text searching.