Object Oriented Database Learning Center

Objectivity Home - Government - Objectivity/DB - Webinars - Download Software

Download a PDF version of this Article

Learning Center Home

Table of contents:

Objectivity/DB - Flexible Deployment

• Introduction
• Single Logical View
• Simple Distributed Servers
• Replaceable Components
• Configuration Parameters
• Processor Power
• Data Locality
• Input/Output Capabilities
• Network Bandwidth
• Concurrency
• Caching
• Data Usage Dynamics
• System Latency
• Fault Tolerance
• Geography/Mobility
• Security
• Configuration Examples

Data Locality

In a central server database environment all data is transmitted to or requested from a single logical server node. Objectivity/DB is a distributed DBMS, so data can be captured, processed, cached and accessed at any convenient location.

For instance, raw data collected from an instrument or sensor can be stored within or close to the equipment that created it, such as near the factory floor or in a moving vehicle. Initial processing may occur there or in separate, probably larger, systems that read the data, correlate it with other data resources and create a repository of more useful information. The original data may stay where it is or be moved to the information repository, depending on usage patterns and local resources.

The processed data may be accessed from multiple sites, some of which may need to reach back to the raw data, or it may be partially or entirely replicated to other sites for further processing, or to alleviate network bandwidth problems. Some subsets of the data may be moved to mobile computers, supplemented with new information and returned to the repository when a connection is available. Figure 4 (below) summarizes this commonly encountered data locality pattern.

Deciding where to store individual items or groups of data depends on many of the other parameters mentioned in this section, particularly: processor power; I/O and network bandwidth; data usage patterns; fault tolerance and security. Objectivity/DB federations can be logically and physically segmented using the database and container constructs. This topic is discussed in depth in a separate White Paper – “Dynamic Containers – the key to superior performance”, discusses this topic in depth.