Hi, 'threads' are nested in a 'forum', therefore, I decided to create a column-family 'thread' with a column named 'parent'.
Is this idea matched with Cassandra philosophy? Because I feel that extracting the data recursively is computational hungry for a site like reddit.com. Therefore one idea is to put the complete data in a JSON and store it in a single column. I hope the retrieval would be fast but the only glitch is; writing/inserting the data back to the Cassandra. Would it be time consuming? -- Thanks & Best Regards, Waqar Azeem
