As databases get larger and larger, managing them becomes more difficult. Websites are regularly creating terabits of data every day, and real-time management for big data requires adapting to the new technologies creating that data.
ManageEngine (News - Alert) is a real-time IT management company, and it recently announced that it is starting to support NoSQL applications Cassandra and MongoDB. NoSQL is designed to accommodate distributed data stores with very large scale data needs.
A Relational Database Management System (RDBMS) scales up by accelerating hardware and using more memory. NoSQL, on the other hand, can do what is called “scaling out,” a process which allows it to spread the load over many commodity systems.
Accomplishing the task in this manner makes it a less expensive solution for large databases that are trying to save money wherever they can.
The fact that the solution is cheaper, however, doesn’t mean it’s less complex to manage. ManageEngine uses its Application Manager to monitor the performance of Cassandra and MongoDB databases to make sure they’re working properly.
This includes memory utilization, database connections, memory allocation and complete monitoring.
In the case of Cassandra, there is performance monitoring and administration of all nodes. In order to track the high availability of those nodes, there has to be visibility to the database and the distributed node structure.
There’s also performance metrics of compaction count, cache hit rate, garbage collection, heap size and usage and IO wait.
When monitoring for MongoDB, the Application Manager keeps track of CPU usage, database operation performance and latency, lock current queue, journaling statistics, memory utilization details, response time and transactional details.
"NoSQL technologies are relatively new and expertise is hard to find, and that makes it difficult to ensure the performance of big data. With out-of-the-box support for Apache Cassandra and MongoDB, Applications Manager can empower existing IT teams and make it easy for businesses to justify their investments in big data technologies," said Gibu Mathew, director of product management at ManageEngine.
As the power of mobile devices and the speed of broadband increases, data will be created at an even more frightening pace. NoSQL may be suitable for now, but better solutions have to be created to adequately manage the immense amount of data that is being created not only by people, but by machines as well.
Edited by Braden Becker