NoSQL databases like Cassandra and Couchbase are quickly becoming key components of the modern IT infrastructure. However this modernization creates new challenges – especially for storage. These modern applications all count on DRAM memory to deliver rapid results to user queries, but while memory is fast it is also expensive and there is only so much that can be installed per server.
When these applications run out of RAM they have to fetch data from storage devices and when that happens performance declines…rapidly. There is also the challenge that memory is volatile, new data must be quickly written to persistent storage.
The Problems with NoSQL Architectures
The default design of a NoSQL architecture, like those Cassandra and Couchbase use, leverage commodity. However, white box servers to keep costs down. They cluster servers together creating a scale out architecture that handles both compute and storage needs. In those servers, IT designers…
View original post 274 more words