
*** check issues marked in ClusterjLoad against latest Cluster/J version
    - leaking connections ?

    + use blind update for setByPk, setB0ToA, nullB0ToA (can be batched)
      and String updates

    + getByPK_[batched] + string reads: define a multi-range index scan
      use an 'in/or' query, then iterate result

    + navA->B0: do not find the As

*** Operations.cpp: remove #ifndef NDEBUG... to ensure consistent
    result verifications across all operations and backends

*** add configuration to run OpenJPA on Cluster/J

*** integrate CRUND benchmark into ... suite

*** add Blob/Text read/write operations to all *Load classes

*** extend JdbcLoad, JpaLoad, ClusterjLoad for Varchar/Varbinary operations

*** tune for OpenJPA performance:
    http://openjpa.apache.org/builds/latest/docs/manual/manual.html#ref_guide_optimization

*** Add/fix support for string length > 100 

*** Review code/patterns/lock modes 

*** Implement patterns: 
    NdbRecord 
    (Asynchronous)
    Investigate NDBAPI anomalies 

*** Fix navA->B0 (NdbApiLoad) ?


