STINGER – taken from the STINGER site
Many problems today can be formulated as dynamic spatio-temporal graph problems. For example, one may wish to track communities within social networks on Facebook as edges (friendship pairs) are added or removed. Or more interestingly, one may look for people who bridge between different social communities, or switch allegiances over time.
As the computer science community increases its development of algorithms and codes for large-scale graph problems, no canonical graph representation has yet to emerge. Without a standard graph representation, algorithms that are implemented for one framework may require substantial programming efforts to port to a different framework. Even worse, algorithms within a single framework may use different data structures for each graph kernel, requiring costly data transformations between each graph kernel subroutine. These inefficiencies in time, space, and productivity, could be reduced or eliminated through a canonical graph representation.
- Portability: Algorithms written for STINGER can easily be translated/ported between multiple languages and frameworks
- Productivity: STINGER should provide a common abstract data structure such that the large graph community can quickly leverage each others’ research developments. This is similar in philosophy to the numerical algorithms community implicit use of sparse and dense matrices.
- Performance: It is recognized that no single data structure is optimal for every graph algorithm. The objective of STINGER is to configure a sensible data structure that can run most algorithms well. There should be no significant performance reduction for using STINGER when compared with another general data structure across a broad set of typical graph algorithms. STINGER should assume a shared memory address space, and allow both sequential or parallel algorithms. The data structure should allow parallel algorithms to exploit concurrency where possible.
Leave a Reply