Designing the Data Structure › Choose to Distribute or Replicate Data › Challenges of Distributed and Replicated Data › Performance

Performance

It is likely that the system design will need tuning for optimum performance. The precise nature of this tuning depends on the individual system environment.

Database performance considerations:

• Is database performance affected by the location of the data?
• If the data is distributed, is access to the data fast enough across the network?
• If the data is centralized, can the server handle the anticipated number of transactions?
• If necessary, consider using distributed data or replicated data.
• Is the performance of the network adequate?
• This can be assessed by looking at the number and size of data transfers. For instance, would a small number of large data transfers or a large number of small data transfers provide better performance?
• How many concurrent clients are there likely to be?
• If there are a large number of users accessing a single data source, will there be problems with users being locked out of the data?
• Is there a potential bottleneck at the communications bridge or server?
• Must the system be available at all times? Can the system cope with machine failures?
• If the workstations are considered inadequate for backup, a more controlled approach is needed.