GFlink: An In-Memory Computing Architecture on Heterogeneous CPU-GPU Clusters for Big Data.
Cen ChenKenli LiAijia OuyangZeng ZengKeqin LiPublished in: IEEE Trans. Parallel Distributed Syst. (2018)
Keyphrases
- big data
- heterogeneous computing
- compute intensive
- multithreading
- memory bandwidth
- memory access
- high volume
- level parallelism
- memory hierarchy
- commodity hardware
- cloud computing
- social computing
- graphics processors
- data analysis
- data processing
- parallel computing
- graphic processing unit
- big data analytics
- real time
- intel xeon
- data management
- social media
- gpu implementation
- graphics processing units
- data warehousing
- vast amounts of data
- unstructured data
- data mining
- data transfer
- business intelligence
- main memory
- management system
- data science
- single instruction multiple data
- knowledge discovery
- information retrieval
- massively parallel
- shared memory
- parallel processing
- cache misses
- information processing
- machine learning
- general purpose computing
- relational databases
- decision support
- memory management
- parallel computation
- parallel programming