Introduction to Java Grid
Java Grid Implementation
Nowadays, the capability of processing massive transactions is required to most of enterprises.
There are two most popular products in java grid at my writing. GigaSpaces and Oracle Coherence.
Both of those products have good features of linear scalability and low latency which can not be accomplished easily with other way, for instance, MQ.
Such single point of failure / bottleneck which most of DBMS have problem with can be also resolved with these products.
With master-worker pattern support, a lot of jobs can be splitted into the individual worker application, processed, and finally the results
will be returned back with very easy way.
Let's look at some usages of Java Grid in next. I will show them with GigaSpaces.
Usage: Logging Service
- The individual application writes log data object into the space which is in this case used as a message bus.
- Logging service will take all the log data with event-driven manner.
- The log data catched will be saved into the file system.
Usage: Real Time Monitoring
- In the similar way to logging service above, the individual application writes log data(common log or exception, error log) into the space.
- Monitoring Service will take all the logs.
- The states of specified monitoring elements will be shown in the dashboards.
- For instance, if some elements status exceed the threshold specified for it, they will be forwarded to the alerts(emailing, SMS, etc).
Usage: Parallel Processing
- Master Service dispatches the job messages into the space.
- Worker Service will detect the job message which it is interested in.
- Worker will process the job message and return back to the space.
- Finally Master service aggregates all the results processed in the space.
References
page_revision: 4, last_edited: 1208351692|%e %b %Y, %H:%M %Z (%O ago)
