How to optimize cfquery

Question

i am facing few problems in a search page where we use
CFquery my cfquery is given below i tried with CF_supercache but it
failed . Is there any other way to optimize the query.
<cfquery name="rsCount" datasource="eservice_reporting"
blockfactor="1000">
select
count(1) as COUNT from kb_sessions s
join REGION r on s.xcntry = r.xcntry and s.xlang = r.xlang
join FAMILY f on s.prodid = f.prodid
where stime between to_date('#bDate#','mm/dd/yyyy') and
to_date('#eDate#','mm/dd/yyyy hh24:mi:ss')
and exitcond LIKE '%'
<cfif selColumns[1] and session.repRegion NEQ 'All'>and
region = '#session.repRegion#'</cfif>
<cfif selColumns[2] and session.repCountry NEQ
'All'>and s.xcntry = '#session.repCountry#'</cfif>
<cfif selColumns[3] and session.repLang NEQ 'All'>and
s.xlang = '#session.repLang#'</cfif>
<cfif selColumns[4] and session.repFamily NEQ 'All'>and
f.family = '#session.repFamily#'</cfif>
<cfif selColumns[5] and session.repProduct NEQ
'All'>and s.PRODID = '#session.repProduct#'</cfif>
<cfif selColumns[6] and session.repUserType NEQ
'All'>and stype = '#session.repUserType#'</cfif>
<cfif selColumns[7] and session.repEmpID NEQ '' and
session.repEmpID NEQ 'All'>and empid =
'#session.repEmpID#'</cfif>
group by
<cfif
selColumns[8]>to_char(stime,'yyyy-mm-dd'),</cfif>
<cfif selColumns[1]>region,</cfif>
<cfif selColumns[2]>country,</cfif>
<cfif selColumns[3]>s.xlang,</cfif>
<cfif selColumns[4]>family,</cfif>
<cfif selColumns[5]>s.prodid,product,</cfif>
<cfif selColumns[6]>stype,</cfif>
<cfif selColumns[7]>empid,</cfif>
exitcond
</cfquery>
any method is there to optimize cfquery without changing it.
with regards,
kaushik:confused;

Answer 1

Optimization you can do without changing the query:
Add
cachedwithin=#CreateTimeSpan (0, 0, 2, 0)# to your cfquery
statement -- although this may not help much due to the nature of
your query.
In the DB, make sure there are indices on:
country,
empid,
exitcond,
family,
region,
s.prodid,
s.xlang, and
stype.
If you didn't have indices, adding them can provide a
dramatic boost.
Otherwise, for big performance gains you are going to have to
make some changes:
This, "s.prodid,product" in the order by is redundant. At best,
high end DB's will optimize it to "s.prodid".
At worst, it will slow the query without changing the
results.
Change it to "s.prodid".
At best "exitcond LIKE '%'" will be optimized out. At worst, it
will force the DB to ignore your indices! Remove this statement.
Do not use "Count" as an alias! This is a reserved word.
Since you are using "yyyy-mm-dd", create a column that contains
"yyyy-mm-dd" and create an index on that column.
to_char can really bog down queries.
Depending on your DB, you can create the new column in such a
way that it sets itself automatically from stime.
Finally, every column in an order by clause sucks down a lot of
processing. Eliminate any column that you can.