Big Data Evaluator Homepage

BDEv Features

BDEv supports the following features, which are executed in an automatic way.
  • BDEv performs:
    • Automatic configuration of the frameworks
    • Deployment of the frameworks using different cluster sizes
    • Execution of predefined and user-defined workloads (batch and interactive modes)
  • BDEv automatically records:
    • Output and elapsed time of the workloads
    • Performance and scalability results
    • Configuration and log directories of the frameworks
    • Resource utilization stats
    • CPU power consumption stats using RAPL
    • Microachitecture-level events using Oprofile
    • Energy consumption stats using HP iLO technology
    • Automatically-generated graphs
  • BDEv allows:
    • To execute the workloads using different frameworks and cluster sizes
    • To unify the configuration of the different frameworks
    • To configure timeouts for the workloads
    • The use of high-performance resources like IP over InfiniBand
    • The use of multiple disks to store intermediate data for the frameworks

Supported frameworks as of BDEv 3.0

Framework Version Deploy mode Network interfaces
Hadoop 1.2.1 JobTracker GbE / IPoIB
Hadoop-YARN 2.7.4 / 2.8.1 YARN GbE / IPoIB
Hadoop-UDA 1.2.1 JobTracker IPoIB
Hadoop-UDA-YARN 2.7.4 / 2.8.1 YARN IPoIB
Flame-MR 0.10.0 / 1.0 YARN GbE / IPoIB
RDMA-Hadoop 0.9.9 JobTracker GbE / IPoIB
RDMA-Hadoop-2 1.1.0 / 1.2.0 YARN GbE / IPoIB
DataMPI 0.6.0 Standalone GbE / IPoIB
Spark 1.6.3 / 2.2.0 Standalone / YARN GbE / IPoIB
RDMA-Spark 0.9.3 / 0.9.4 Standalone / YARN GbE / IPoIB
Flink 1.2.1 / 1.3.2 Standalone / YARN GbE / IPoIB

Supported benchmarks

Benchmark Input generator Source code
Hadoop DataMPI Spark Flink
Testdfsio - Hadoop examples - - -
Wordcount RandomTextWriter (Hadoop) Hadoop examples DataMPI examples BDEv* BDEv*
Sort RandomTextWriter (Hadoop) Hadoop examples DataMPI examples BDEv* BDEv*
Grep RandomTextWriter (Hadoop) Hadoop examples DataMPI examples BDEv* BDEv*
Terasort TeraGen (Hadoop) Hadoop examples DataMPI examples BDEv* BDEv*
Pagerank DataGen (HiBench) Pegasus - BDEv* BDEv*
Connected Components DataGen (HiBench) Pegasus - BDEv* BDEv*
Bayes DataGen (HiBench) Apache Mahout - BDEv* -
Kmeans GenKMeansDataset (Mahout) Apache Mahout - BDEv* BDEv*
Aggregation DataGen (HiBench) Apache Hive - BDEv* -
Join DataGen (HiBench) Apache Hive - BDEv* -
Scan DataGen (HiBench) Apache Hive - BDEv* -
Command User-defined User-defined User-defined User-defined User-defined
*Adapted from Spark and Flink examples when possible