Various parts of the code might retrieve a node or edge from disk repeatedly when they could just retrieve it once.
Right now, a DiskNodeRef will copy over all the data from the node, and we might only be accessing one field of it, then freeing it, and creating a new DiskNodeRef to access a different field. Accessing both fields from the same reference would avoid another read/write.
Somewhat related to #61.
Various parts of the code might retrieve a node or edge from disk repeatedly when they could just retrieve it once.
Right now, a
DiskNodeRefwill copy over all the data from the node, and we might only be accessing one field of it, then freeing it, and creating a newDiskNodeRefto access a different field. Accessing both fields from the same reference would avoid another read/write.Somewhat related to #61.