Database Reference
In-Depth Information
Implementing a benchmarking framework
on multiple instances
Benchmarking a platform is highly essential in order to keep up the Service Level
Agreements (SLA) for your application and database environment. Also, the task will enable
the service orientation behavior as a proactive manner that will avoid any unprecedented
activity on the data platform.
The key result out of benchmarking is to find the problem with a consistent view of services,
and familiarize with application operational behavior to streamline any issues that are
persistent, not to mention about performance monitoring and troubleshooting. There are no
technical details on how to implement this, but in this recipe we will go through a tool that will
help in implementing a benchmarking framework on multiple instances of the data platform.
TPC
stands for
Transaction Processing Council
, which
is a non-profit corporation founded to define transaction
processing and database benchmarks to disseminate
objective, verifiable TPC performance data to the industry.
Getting ready
The essential benchmarking requirements that are inline with SLAs are as follows:
F
In the event of any failure, the database environment must be available within X
minutes
F
Data for the last six months must be available online
F
The backups for the last eight years must be stored offsite and must be available
within X hours during any requirement
F
In the next two years, database size is expected to be (capacity)
F
Obtain industry standards test such as TPC-C or TPC-H
How to do it...
In case the database platform is a highly transactional OLTP system then it is required to test
the hardware and systems as per the industry standard TPC-C.
Every hardware vendor has his own standards implementation at his end; however, it is
performed as high-level testing.
Search WWH ::
Custom Search