hadoop - Will reducer out of java heap space -


i implementing program using hadoop. question how deal java out of space problem, added property configuration xml file, didn't work. increasing number of reducers doesn't work me either. because in program every reducer needs large sparse whole matrix, , not allowed change logic. yet every reducer receive entry column id key, , column vector value.is there way can out of dilemma?

ps:i first thinking reducer take column 1 one, , won't cause out of memory issue. seems not way. actually, trying implement algorithm described in paper. algorithm reducer listed below. line 11 column mentioned, , each reducer has every column of matrix. enter image description here

the reason why use hadoop because cant fit entire data set memory. either don't change logic , try find computer that's big enough or parallel-ize algorithm , exploit hadoop.


Comments

Popular posts from this blog

c# - Send Image in Json : 400 Bad request -

jquery - Fancybox - apply a function to several elements -

An easy way to program an Android keyboard layout app -