Reaching required capacity
A large software company was developing a WEB2.0 product and initial performance tests showed major performance
degradation when
reaching 20% of capacity. Solving the problem was a matter of taking several steps - none overly
complex. These steps allowed the product to reach planned capacity.
Adding DB indexes to some tables, changing the way entities were stored
inside LDAP repository gave the final boost which let performance rock.
Releasing Memory Pressure
Performance analysis methodology was applied to the monitoring product
developed by a large software company. One of the major challenges with respect to the product was its memory
footprint. While
changing the data model and object representation could reduce memory
footprint by an order of magnitude, this was not a valid option due to
project constraints. After analyzing memory consumption, we were able to
reduce memory footprint by 30% through a very simple change that
was so safe, it was included in the next version as well as the current
version's service pack. Since memory foot-print was the biggest
bottleneck, this simple solution increased product capacity and solved many customer escalations.
References are available upon request.
| 