Hi all, what do you think about creating a branch for each scalability improvement we aim to test? This way we wouldn't pollute our trunk but then also others could try out the work we're doing on such a scalability "feature". So for example I'd create branch-732 for investigating the usage of Apache DirectMemory for storing graph vertices, then if at some point it looks nice that could be merged back to trunk and the same for other features.
What do you think? Tommaso
