sql - Horizontal scaling of search query -


we building cv scoring service, , using postgres making complex queries find cv's match vacancy best. problem is, use complex set of heuristics score cv vacancy, , average number of cvs scored per query growing. want put kind of load outside of database, , looking existing solutions horizontal scaling such load. query should executed in fraction of second, there can hundreds of concurrent queries. each query scores on average 10k cvs. each cv 50 records in maybe 10 tables in current relational form.

i want clustered system run each query in multiple parallel processes (on many servers) , return aggregated result. should fast , fault tolerant. looking hadoop, looks designed batch processing, , not realtime low latency load. there apache storm, designed continous stream processing. not shure :)

what kind of tool suit needs? thank you!

  1. make sure not redoing work, if cv has been scored tag scored , don't reprocess unless it's necessary.

  2. unless partitioning data in postgres might want that. not rows need accessed regularly.

  3. sounds want scale reads, in case postgres read-only cluster option.

  4. take @ elasticsearch, designed weighted scoring, faceting, etc. should scale, haven't tried myself though.

i start 1 though, don't work unless have to.


Comments

Popular posts from this blog

c++ - Difference between pre and post decrement in recursive function argument -

php - Nothing but 'run(); ' when browsing to my local project, how do I fix this? -

php - How can I echo out this array? -