> Compacted row maximum size: 36904729268 So 36 gigs. As long as you're sure each column is only about 1k, the total row size should not be a problem.
> While I don't see OOMs when I use only a single thread to page the row, there > are lots of ParNew collections that take about 500ms each and also many full > collections. > > Do I just not have enough RAM? Unless something is wrong, I suspect the problem is the total amount of rows that you have which affects the size of bloom filters. I'm a bit out of date and don't remember how the memory sizing calculations are currently made. I suspect you either have to simply add heap size, or decrease memtable sizing. I would attach with jconsole and l look at the memory usage after each concurrent mark sweet phase (or just hit "perform gc" and watch mem usage). That will tell you the live size. It is likely close to filling your heap. I did a quick calculation and each 2 billion column with RF=2 should result in bloom filter sizes of ~ 3.7 gig per node... I probably am overlooking something though since you say you have 4 GB of RAM, unless you have your heap size set to almost 4 gig? -- / Peter Schuller (@scode, http://worldmodscode.wordpress.com)
