SAS Visual Analytics can
load data from registered tables or SASHDAT files stored in HDFS to
the SAS LASR Analytic Server. The following table identifies
some of the advantages and disadvantages of each method:
Comparison of Registered Tables and HDFS Data Sources
|
|
|
|
Registered table (from
a library that does not use a co-located data provider)
|
Provides a rapid method
for loading tables.
|
Appropriate for smaller
data sets because the data must be transferred over the network.
If the table is unloaded
or the server stops, the data must be transferred over the network
again.
|
SASHDAT file in HDFS
or a registered table from a library that uses a co-located data provider
|
Impressive performance
for loading very large data sets in parallel.
If a SAS LASR Analytic Server
is stopped or the table is unloaded from the server, loading it again
is also very fast.
|
Requires a separate
step to add the data to HDFS or the co-located third-party vendor
database and then load it before the data is available to SAS clients.
|