HashGraph—Scalable Hash Tables Using a Sparse Graph Data Structure

In this article, we introduce HashGraph, a new scalable approach for building hash tables that uses concepts taken from sparse graph representations—hence, the name HashGraph. HashGraph introduces a new way to deal with hash-collisions that does not use “open-addressing” or “separate-chaining,” yet it has the benefits of both these approaches. HashGraph currently works for static inputs. Recent progress with dynamic graph data structures suggests that HashGraph might be extendable to dynamic inputs as well. We show that HashGraph can deal with a large number of hash values per entry without loss of performance. Last, we show a new querying algorithm for value lookups. We experimentally compare HashGraph to several state-of-the-art implementations and find that it outperforms them on average 2× when the inputs are unique and by as much as 40× when the input contains duplicates. The implementation of HashGraph in this article is for NVIDIA GPUs. HashGraph can build a hash table at a rate of 2.5 billion keys per second on a NVIDIA GV100 GPU and can query at nearly the same rate.

EFFICIENT DESIGN OF STORAGE AND RETRIEVAL METHODS

The paper demonstrates the viability of a number of storage techniques such as hashing using multilingual texts. The empirical studies of hashing techniques based on multilingual provide researchers with insight information into the structure of the methods and their applicability. This may lead to more theoretical findings to enhancing performance of storage and retrieval algorithms. A number of experiments were performed using real and random texts from English and Arabic languages with numerous sets of data items and length sizes. The English language is used in the experiments as measurement tools as it is the international language of the world and it has been received the best attention from researchers all over the world. The study shows that chaining techniques consistently generates less number of collisions than open addressing techniques for texts from any language. The motivation behind conducting this study was the lack of research on the performance of hashing algorithms using data items that have dependencies and structures from other than English language. There are also no comprehensive studies performed on Arabic strings. The new finding probably leads to design more efficient hashing algorithms for storing, deleting and retrieving items of information.