Gram: Scaling Graph Computation To The Trillions

GRAM is an efficient and scalable graph engine for a large class of widely used graph algorithms. It is designed to scale up to multicores on a single server, as well as scale out to multiple servers in a cluster, offering significant, often over an order-of-magnitude, improvement over existing distributed graph engines on evaluated graph algorithms. GRAM is also capable of processing graphs that are significantly larger than previously reported. In particular, using 64 servers (1,024 physical cores), it performs a PageRank iteration in 140 seconds on a synthetic graph with over one trillion edges, setting a new milestone for graph engines.GRAM's efficiency and scalability comes from a judicious architectural design that exploits the benefits of multi-core and RDMA. GRAM uses a simple message-passing based scaling architecture for both scaling up and scaling out to expose inherent parallelism. It further benefits from a specially designed multi-core aware RDMA-based communication stack that preserves parallelism in a balanced way and allows overlapping of communication and computation. A high degree of parallelism often comes at the cost of lower efficiency due to resource fragmentation. GRAM is equipped with an adaptive mechanism that evaluates the cost and benefit of parallelism to decide the appropriate configuration. Combined, these mechanisms allow GRAM to scale up and out with high efficiency.