Apache Giraph Reviews

Graph Engine (GE), a distributed in-memory processing engine, is supported by a strongly-typed RAM storage and a general distributed computing engine. The distributed RAM store is a global addressable, high-performance key-value storage that can be accessed by a cluster of computers. GE's RAM store allows fast random data access over a large data set. GE is a natural platform for large graph processing due to its ability to speed data exploration and distribute parallel computing. GE supports both low latency online query processing as well as high-throughput offline analysis on billion-node large Graphs. Schema is important when data processing must be efficient. For data storage that is compact, quick and clear, strong data modeling is essential. GE has the ability to manage billions of runtime objects of different sizes. As the number of objects increases, each byte counts. GE offers fast memory reallocation and allocation with high memory ratios.

PuppyGraph allows you to query multiple data stores in a single graph model. Graph databases can be expensive, require months of setup, and require a dedicated team. Traditional graph databases struggle to handle data beyond 100GB and can take hours to run queries with multiple hops. A separate graph database complicates architecture with fragile ETLs, and increases your total cost ownership (TCO). Connect to any data source, anywhere. Cross-cloud and cross region graph analytics. No ETLs are required, nor is data replication. PuppyGraph allows you to query data as a graph directly from your data lakes and warehouses. This eliminates the need for time-consuming ETL processes that are required with a traditional graph databases setup. No more data delays or failed ETL processes. PuppyGraph eliminates graph scaling issues by separating computation from storage.