Dear All, Please help to provide input to design requirement sent in previous mail. is there any way to search past months archive "lucene-solr-user mailing list archives"?
is it possible to achieve below solution to merge result from Solr and mysql? How to search if one field exist in solr and other in mysql? E.g-Suppose we have indexed product information (changing not frequently e.g-Brand,size,location etc.) in Solr and quantity & price in mysql as this data change frequently (one document for each store). 1.if user select brand then we can search directly from Solr first on the basis of brand & user location and then get price and quantity from mysql on the basis of ids retrive from solr (maximum 50 product per page). This seems to be fine keeping in view performance. 2.if user select price and brand then in this case first we need to search mysql on the basis of price range to retrieve product ids.These product ids need to pass to solr to filter on the basis of brand and location.This might not be good idea because ids may be in million as well.Please suggest what is the best way to handle these scenarios? Looking forward for your help. Thank you.. Sumit Agarwal On Mon, Feb 22, 2016 at 6:35 PM, Sumit Agarwal <[email protected]> wrote: > Dear All,. > > > Please share your input to design schema on the basis of below input? > > > How to design Solr schema in the case of Multi store marketplace > application like Amazon? > > > I have asked to design solr schema for application in which product is > selling by multiple store and each store is having its own quantity and > price. Keeping in view this requirement > > I can think below schema design. > > > 1. Demoralize data for store and product table. In this case there is > lot of redundancy and if 7m product sold by 100 stores then it will be > 100X7M record. I don’t think good design because it will have lot of > redundancy in each document and 7M documents will become 700M documents in > solr. Apart from this On Product listing page I need to show product along > with minimum price exist among the stores. To get this information I need > to get group store products and pass to group query as price ascending so > that minimum price document will be retrieved per group. As per Google > Group performance with facet is bad in solr. > > 2. If I kept product document separately from store documents then I > need to execute more than a query to get products attribute and store > product having minimum price. This may be performance issue. > > 3. Solr has introduced new functionality indexing block. As price and > quantity will change frequently then it is difficult to modify these fields > in block because whole block needs to re-indexed every time. > > 4.how to handle if fields such as seller price,offer price are changing > frequently? > > Please suggest what is the best practice to design schema in this scenario? >
