Testing Distributed Loader

Set the number of goroutines for gentle merging to 18. It is little slower to run these goroutines, but they keep the memory usage in check.

Store any errors encountered while running the pipeline. If any error is found, stop processing immediately. Switch the system to sync.WaitGroup completely, instead of an error channel. Log fatal on error in loader and uidassigner

Loader shouldn't cause uid assignment. Add a Get() function in uid, and avoid calling rdf.GetUid. Plus, other changes across the entire code base to accommodate this.

Change how many keys we merge. Also, run gentlyMerge in a goroutine, so our memory monitor can call aggressivelyEvict in time. Also, with this change, the memory growth is a heck of a lot more controlled.

use new Lexer initializer

Distributed query and mutation testing

Create a sync.Pool for entry struct. This removes newOrExisting in uid/assigner.go from memory profiler.

Allocate lexer on stack instead of heap. This now makes rdf.Parse very cheap, it no longer shows up in top10.

Add memory profiling. Use a sync.Pool for lexer. Add benchmark numbers before and after using sync.Pool. Simplify assignUids only.

Handle absense of mutation or query in GraphQL. Fix bugs in goroutine's variable access, and pointer assignment to mutationArray.

Distributed Mutations: Xid to Uid, NQuads to Edges, Apply Edges

Assign uids for xids over the network. Convert NQuads to Edges using the assigned uids, and then apply these as mutations over the network.

Distributed mutations from DirectedEdges

Distributed server