GPA Manual for G-series RVUs

Guardian Performance Analyzer (GPA) Manual – (544977-001) Page 59 of 106
Section 4: Using GPA Information
Interpreting the Standard Report
A GPA analysis can detect and point to a number of causes of poor or inefficient
system performance. For some of the problems, such as a memory shortage or a load
imbalance, GPA can make appropriate tuning recommendations and even, in some
cases, provide the means for automatically implementing the recommendations. For
other performance issues, such as transient processing, GPA can reveal the condition,
but the user must decide how to correct it. (For a general discussion of system tuning,
see Section 4 in the Measure User’s Guide.)
This subsection discusses the various sections of the GPA Standard Report described
in Section 3: Description of GPA Reports. In the discussion, we look at three GPA
analysis examples. One is for a node (\NODEA) in which the main problem is a serious
shortage of memory, one is for a node (\NODEB) that exhibits a CPU load imbalance,
and one is for a node (\NODEC) that requires a change in the configuration of its
primary disk processes.
With regard to these examples, keep in mind that GPA considers a shortage of
memory to be the most serious problem on a node. If the shortage is severe enough,
as is the case for \NODEA, GPA deals initially only with this issue even though the
analysis may find other tuning problems on the node. Once the memory shortage
problem is corrected, another GPA analysis would consider and make tuning
recommendations for the remaining problems.
System Performance Summary
The System Performance Summary gives you an overall picture of the system and
indicates, in general, how well or how poorly the system analyzed is performing. From
this section of the report, you can do the following:
• Become familiar with the system’s characteristics.
• Assess the general health of the system.
• Spot quickly where any problems may lie.
• Determine the potential for improved system performance.