back to the
main page
Interview: Inside Oracle Application Grid Computing
Application grid computing is promising to improve the efficiency, scalability, and quality of service (QoS) of a wide variety of applications. Oracle Information InDepth editor Jeff Erickson spoke with Ruma Sanyal, director of Product Marketing, to see how it works.
| Editor: |
How is application grid computing different from standard application management techniques? |
| Sanyal: |
Traditionally, an application is resourced with its own individual stack, which is both wasteful and limiting. Each application must be provisioned with capacity for its worst case, meaning that most of the time this capacity is unused. And it’s limiting because the underlying server’s processing power, memory capacity, and storage capacity form an upper bound of the application’s performance.
For example, if a particular application required memory beyond the server’s physical memory, it had to move data between memory and disk and work on subsets, a process that slows performance.
With the resource sharing and automated management provided by Oracle Fusion Middleware technologies in an application grid, your physical resources are much more efficiently utilized.
|
| Editor: |
What Oracle products come together to create an application grid? |
| Sanyal: |
To make application grid computing possible, Oracle brings together several industry-leading technologies: Oracle WebLogic Server for the Java domain, Oracle Tuxedo for the C/C++/COBOL domain, Oracle Coherence in-memory data grid, Oracle JRockit Java runtime solutions, and Enterprise Manager with WebLogic Operations Control for automated management. |
| Editor: |
How does an application grid improve scalability? |
| Sanyal: |
In two fundamental ways. First, with resource sharing, scale becomes virtually unlimited. By introducing an in-memory data grid, for example, a single application can draw on the fast memory of multiple machines, providing virtually unlimited computing power. An application can now draw upon much broader resources than a single server could provide, enabling much higher reliability and quality of service. Second, the application grid enables you to more optimally vary the scale by adding and removing resources dynamically. With the in-memory grid, you add and subtract nodes as your demand varies. |
| Editor: |
Tell us more about Oracle Coherence in-memory data grid. |
| Sanyal: |
Oracle Coherence is the premier distributed, in-memory data grid, scaling linearly up to thousands of servers. It automatically and dynamically partitions data in memory across multiple servers, enabling continuous data availability and transactional integrity, even in the event of a server failure. This enables both higher performance and higher reliability, which together mean higher quality of service for the end user. |
| Editor: |
How do you manage the application grid? |
| Sanyal: |
Enterprise Manager with WebLogic Operations Control automates resource provisioning that adapts dynamically to applications needs. The management layer also provides predictive alerts, historical data, and root cause analysis for all aspects of the application, so operators can rapidly analyze and improve performance levels. |
Learn more about Oracle Application Grid computing:
back to the top
|
