Self-Balancing Job Parallelism and Throughput in Hadoop.
Bo ZhangFilip KrikavaRomain RouvoyLionel SeinturierPublished in: DAIS (2016)
Keyphrases
- response time
- open source
- commodity hardware
- parallel processing
- map reduce
- parallel computation
- memory bandwidth
- cloud computing
- low latency
- massively parallel
- data intensive
- single machine scheduling problem
- data sets
- shared memory
- distributed computing
- data management
- distributed systems
- case study
- load balancing
- resource utilization
- quality of service
- parallel computers
- completion times
- parallel architectures
- fine grain
- level parallelism