Cassandra draws heavily from BigTable (data model) and Dinamo (architecture), two databases doorbell installation of today's most popular and powerful. That alone may be enough to consider it, but I'll go ahead and provide the list of questions you can ask yourself:
Cassandra provides fantastic doorbell installation recording and very good read throughput (beating most popular competitors), comparable only to HBase. Deciding between the two is largely a matter of personal preference. If you do not want the entire Hadoop stack with all its moving parts and increased the complexity of the architecture, choose Cassandra, Hadoop is good, HBase can be integrated better fit. If you are dealing doorbell installation with small to medium volumes of data, relationship solutions become much more interesting, because of the flexibility they provide.
Cassandra reduces its data model to the absolute minimum in order to keep the partition data set as possible. This pays back when talking about linear scalability, but leaves you with "nothing more" than a distributed hash map. But do not judge too early, as even one of the key technologies of Google Bigtable, is exactly the same. Since C provides a programmer with a fantastic performance in the cost of some comfort features such as garbage collection, Cassandra provides you with a central piece of Google-scale technology, the cost of dealing with indexes of yourself or think about the queries that you need beforehand.
Because doorbell installation of very low level Cassandra data model, applications require extensive knowledge about the dataset. If transparency is the application you need, Cassandra is not for you.
2013 (9) May (4) April (2) Cassandra is the best option for me? TCC project on Cassandra March (1) January (2) 2012 (7) November (3) September (3) May (1)
No comments:
Post a Comment