As the volume and complexity increase, and as we try to improve more complex processes we need more high performance computing (HPC). Its a complex thing to build code that we know will operate correctly and predict well at very large scale.
'A System Purely for Developing High-Performance, Big Data Codes'
Rice News by Jade Boyd
Rice University researchers this week will debut the PlinyCompute platform at the 2018 ACM SIGMOD conference in Houston, TX. Developed under Rice's Pliny Project banner, the PlinyCompute tool was created to ease implementation of complex objects and workflows on big data platforms. Rice's Jia Zou says PlinyCompute differs from the Spark platform "because it was designed for high performance from the ground up." Compared to Spark, benchmarking showed PlinyCompute was at least twice as fast, and in some cases 50 times faster, at implementing complex object manipulation and library-style computations, Zou says. The Pliny project is aimed at creating programming tools that can "autocomplete" and "autocorrect" code. ... "
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment