[ Thanks to E5Rebel
for this link. ]
“1. Portal: Anything but the simplest static HTML page has some
sort of data access module at the back-end that formulates a query
to fetch relevant records or composite views from one or more data
stores. Generally, the overall fetching operations on many NoSQL
technologies could be several orders faster than RDBMS as one can
fetch the data from the shards in a scale-out parallel manner.“But here is where the complication arises: not all NoSQL stores
are equally capable. Out of the box, some do not support range
queries, some support only simple predicates with constant values,
some do not support composite indices or queries that are not based
on keys. Thus, while the data fetch may be blazingly fast, the
complexity is pushed up to the application layer (string-based
joins, anyone?).“Now, if your portal requires data update functionality, the
write operation in NoSQL is generally smooth even in high volume
because many of them exist largely in memory. In contrast,
traditional RDBMS deals with locks, latches, re-do logs and so
forth. However, bear in mind that NoSQL does not generally support
atomicity, consistency and in most cases proper durability. Thus
for applications that require absolute transactional integrity and
serialisation, SQL databases are still the top choice.”