Hi.. Thnk u for ur reply... let me explain my scenario On a sample set od 1000 documents i am generating a matrix of (may be) 1000000 * 1000 .I have much larger document set than this say 50000. So can memcahe be able to store these huge matrix representation as u said??
Sreejith On Thu, Oct 28, 2010 at 6:56 AM, PlumbersStock.com < [email protected]> wrote: > When dealing with large data sets I sometimes split it into parts that > will fit with a master object that stores how many parts it should > have. Of course it could still fall out of cache but it's no worse > than if it was one large object that fell out of cache and possible > better if you kept a primary copy somewhere so you only have to grab > the missing chunks from it. > > big_object: 5 > big_object_0: data > big_object_1: data > big_object_2: data > big_object_3: data > big_object_4: data
