[couchbase2] Update Java SDK version, query for workloade

[kudu] Update Kudu binding to 0.9.0

Shell scripts to run YCSB.

Using RAW queries reduces overhead at cbq-engine on the cluster and
reduces transport.  This in turn delivers lower latencies and thus
higher throughput with YCSB's tight loops of requests.

Adding an ORDER BY maintains strict correctness with the intended
implementation of Workload E.  In this particular case, the
ORDER BY is effectively a noop since the index in use is ordered
anyway.

The README for couchbase2 was also updated to reflect the simpler
index creation which is possible with Couchbase Server 4.5.  Server
4.5 automatically covers on the primary index so there is no need
for the more complicated syntax unless using an older version.

Note this change requires the 2.3.1 Couchbase Java SDK and its
underlying core-io 1.3.1.  The core-io 1.3.1 adds a streaming
parser for RAW results.

Add batching in insertion in JDBC client bindings

Fix a typo in Histogram generator

[mongodb] return OK on updates with w=0

Development of RADOS binding of Ceph FS

[mongodb] added logback.xml to mongodb module

[Release] Update top level README.md file to point to 0.9.0 release.

[elasticsearch] Bump version to 2.3.2

This commit bumps the version of the Elasticsearch dependency for the
Elasticsearch binding to version 2.3.2.

[couchbase2] Better Defaults & Scan Refactor

[riak] Added a workaround to allow strong-consistent scan transactions.

[bigtable] Bigtable uses 1ms granularity.

Fixes #733

[core] Add the Utils.getGCTotalTime() method for showing how long

collector. This will allow greater insight into the generations.
Now the client will emmit the stats for each collector at the end of
run along with the totals.

the JVM spent in GC. It now prints the time and percent of time in
the final output.
Also fix cases where the garbage collector implementation returns
-1 if the count or time isn't implemented.