Beth,

AFAIK Group By is not optimizable, so there is no way. I have this problem myself with statistical queries where I need Group By.

You may try to split it, but it probably would not help.

>Hi All,
>
>does anyone know of a way to make this select statement quicker?
>

>SELECT lnPayRecID AS pay_rec_id,;
>  ldStartDate AS st_dte, ldEndDate AS end_dte,;
>  service_id, cst_frml, cst_coef, ano_frml, ano_coef,;
>  adj_frml, adj_coef, cpi_frml, COUNT(*) AS srv_days,;
>  SUM(IIF(EMPTY(date_calc),0,1)) AS calc_days,;
>  SUM(IIF(EMPTY(date_calc),0.00,NVL(contract_base,0.00))) AS cntr_base ,;
>  SUM(IIF(EMPTY(date_calc),0.00,NVL(contract_anomaly,0.00))) AS cntr_ano,;
>  SUM(IIF(EMPTY(date_calc),0.00,NVL(contract_adjustment,0.00))) AS cntr_adj,;
>  SUM(IIF(EMPTY(date_calc),0.00,NVL(contract_total,0.00))) AS cntr_total,;
>  SUM(IIF(EMPTY(date_appr),0,1)) AS appr_days,;
>  SUM(IIF(EMPTY(date_appr),0.00,NVL(appr_final,0.00))) AS cntr_appr ;
>  FROM cutemp;
>  GROUP BY service_id,cst_frml,cst_coef,ano_frml, ano_coef, adj_frml, ;
>  adj_coef, cpi_frml ;
>  INTO CURSOR new_t
>

>
>Thanks,
>Beth