Please enable JavaScript.
Coggle requires JavaScript to display documents.
Cluster B: Parallel and Distributed Clustering Algorithms (K - Means (K…
Cluster B: Parallel and Distributed Clustering Algorithms
CURE
CURE using OpenMP
[70]
Transparent management of asymmetry, non–determinism in CURE, effective exploitation of irregular nested loop–level parallelism by OpenMPI
Effective scalability
Improved CURE for big data using Hadoop & Mapreduce
[71]
Outliers removal to nullify parameters effect on clustering
Using Oracle 11G
[72]
CRCW PRAM
Speedup & Scaleup
CLARA
Clustering on cloud: CLARA to MapReduce
[73]
Automatic parallelization, fault tolerance on the cloud
Hadoop MapReduce Framework
WaveCluster
Parallel WaveCluster
[67]
Message passing model for distributed-memory multiprocessor system
Superior speedup & linear scaling behavior
Overcomes space complexity constraint due to its distributive nature
Parallel computing of WaveCluster on CUDA
[69]
Speed-up process computing time and improved accuracy of face images
Parallel computing with large memory bandwidth
Parallel wavelet clustering on GPUs using CUDA
[68]
CUDA implementations of WaveCluster and connected component labeling
Divide & conquer wavelet transform, multi-pass sliding window connected component labeling implementation
K - Means
Parallel K-Means with MPI
[75]
High performance, scalability, portability, low time overhead
IPKMeans
[78]
Parallel pre-processing using k-d tree, single MapReduce with more reducers
Lower I/O overheads & fast processing
Parallelizing K-Means on Spark
[76]
Technical barrier & alternative strategies step wise
Effective & efficient implementation
K-Means - CUDA
[74]
Explores trade-off between load balance & memory access coalescing through data layout management
Adaptive triangle inequality determination
Improved speed & scalability
Distributed K-Means on CloudStack
[77]
Hadoop distributed cluster on CloudStack
Good quality clustering, efficient execution time