Estimating Space Required by Clusters Oracle Tips by Burleson Consulting

The procedure given here shows how to estimate the initial amount of space required by a clustered set of tables. This procedure only estimates the initial amount of space required for a cluster. When using these estimates, note that the following factors can affect the accuracy of estimations:

* Trailing nulls are not stored, nor is a length byte. Inserts, updates, and deletes of rows, as well as tables containing rows or columns larger than a single data block, can cause chained row pieces. Therefore, the estimates you derive using the following procedure may tend to be lower than the actual space required if significant row chaining occurs.

* Once you determine an estimate for a table’s size using the following procedure, you should add about 10 to 20 percent additional space to calculate the initial extent size for a working table.

To estimate the space required by clusters, perform the following general steps:

1.    Calculate total block header size and space available for table data.

2.    Calculate the combined average column lengths of the rows per cluster key.

3.    Calculate the average row size of all clustered tables.

4.    Calculate the average cluster blocksize.

5.    Calculate the total number of blocks required for the cluster.

Let’s step through a more detailed example.

Step 1: Calculate Total Block Header Size and Space Available for Table Data

The following formula returns the amount of available space in a block:

space left in block after headers (hspace)

hspace =  BLOCKSIZE -( KCBH - UB4 - KTBBH - (KTBIT*(INITTRANS - 1)) - KDBH)

The sizes of KCBH, KTBBH, KTBIT, KDBH, and UB4 can be obtained by selecting * from v$type_size table.

Then use the following formula to calculate the space available (s_avail) for table data:

s_avail = (hspace*(1 – PCTFREE/100)) – (4*(NTABLES + 1) * ROWSINBLOCK)

where:

BLOCKSIZE. The size of a data block.

INITTRANS. The initial number of transaction entries for the object.

PCTFREE. The percentage of space to reserve in a block for updates.

NTABLES. The number of tables in the cluster.

ROWS IN BLOCK. The number of rows in a block.

Note: Several calculations are required to obtain a final estimate, and several of the constants (indicated by an asterisk (*)) provided are operating system-specific. Your estimates should not significantly differ from actual values. See your operating system–specific Oracle documentation for any substantial deviations from the constants provided in the above procedure.

Step 2: Calculate Space Required by a Row

To calculate this number, use step 3 from the five-step procedure for estimating the space required by nonclustered tables from Chapter 5, Section 5.7, “Sizing an Oracle9i Nonclustered Table,” in. Make note of the following caveats: 

* Calculate the data space required by an average row for each table in the cluster. For example, in a cluster that contains tables t1 and t2, calculate the average row size for both tables.

* Do not include the space required by the cluster key in any of the above data space calculations. Make note of the space required to store an average cluster key value for use in step 5. For example, calculate the data space required by an average row in table t1, not including the space required to store the cluster key.

Do not include any space required by the row header (that is, the length bytes for each column); this space is accounted for in the next step. For example, assume two clustered tables are created with the following statements:

CREATE TABLE t1 (a CHAR(6), b DATE, c NUMBER(9,2))
                CLUSTER t1_t2 (c);
CREATE TABLE t2 (c NUMBER(9,2), d CHAR(6))
                CLUSTER t1_t2 (c);

Notice that the cluster key is column C in each table.

Considering these sample tables, the space required for an average row (D1) of table t1 and the space required for an average row (D2) of table t2 is:

D1 (space/average row)    = (a + b) = (9 + 6) bytes = 15 bytes
D2 (space/average row)    = (d) = 9 bytes

Step 3: Calculate Total Average Row Size

You can calculate the minimum amount of space required by a row in a clustered table according to the following equation:

Sn bytes/row = row header + Fn + Vn + Dn

where:

row header. Four bytes per row of a clustered table.

Fn. Total length bytes of all 250 bytes or less. The number of length bytes required by each column of this type is 1 byte.

Vn. Total length bytes of all columns in table n that store more than 250 bytes. The number of length bytes required by each column of this type is 3 bytes.

Dn. Combined data space of all columns in table n (from step 2).

Tip: Do not include the column length for the cluster key in variables F or V for any table in the cluster. This space is accounted for in step 5.

For example, the total average row size of the clustered tables t1 and t2 is as follows:

   S1 = (4 + (1 * 2) + (3 * 0) + 15) bytes = 21 bytes
   S = (4 + (1 * 1) + (3 * 0) + 9) bytes = 14 bytes

Tip: The absolute minimum row size of a clustered row is 10 bytes and is operating system specific. Therefore, if your calculated value for a table’s total average row size is less than these absolute minimum row sizes, use the minimum value as the average row size in subsequent calculations.

Step 4: Calculate Average Cluster Blocksize

To calculate the average cluster blocksize, first estimate the average number of rows (for all tables) per cluster key. Once this is known, use the following formula to calculate average cluster block size:

avg. cluster block size (bytes) = ((R1*S1) + (R2*S2) + . . . + (Rn*Sn)) + key header
+ Ck + Sk + 2Rt

where:

Rn. The average number of rows in table n associated with a cluster key.

Sn. The average row size in table n (see step 3).

key header. This is 19.

Ck. The column length for the cluster key.

Sk. The space required to store average cluster key value.

Rt. The total number of rows associated with an average cluster key (R1 + R2 . . . + Rn). This accounts for the space required in the data block header for each row in the block.

For example, consider the cluster that contains tables t1 and t2. An average cluster key has one row per table t1 and 20 rows per table t2. Also, the cluster key is of datatype NUMBER (column length is 1 byte), and the average number is 4 digits (3 bytes). Considering this information and the previous results, the average cluster key size is:

SIZE = ((1 * 21) + (20 * 14) + 19 + 1 + 3 + (2 * 21)) bytes = 368 bytes

Specify the estimated SIZE in the SIZE option when you create the cluster with the CREATE CLUSTER command. This specifies the space required to hold an average cluster key and its associated rows; Oracle uses the value of SIZE to limit the number of cluster keys that can be assigned to any given data block. After estimating an average cluster key SIZE, choose a SIZE somewhat larger than the average expected size to account for the space required for cluster keys on the high side of the estimate.

To estimate the number of cluster keys that will fit in a database block, use the following formula, which uses the value you calculated in step 2 for available data space, the number of rows associated with an average cluster key (Rt), and SIZE:

# cluster keys per block = FLOOR(available data space + 2R / SIZE + 2Rt)

For example, with SIZE previously calculated as 400 bytes (calculated as 368 earlier in this step and rounded up), Rt estimated at 21, and available space per data block (from step 2) calculated as 1,742 – 2R bytes, the result is as follows:

# cluster keys per block = FLOOR((1936 – 2R + 2R) / (400 + 2 * 21))
= FLOOR(1936 / 442) = FLOOR(4.4) = 4

Note: FLOOR means round down.

Step 5: Calculate Total Number of Blocks

To calculate the total number of blocks for the cluster, you must estimate the number of cluster keys in the cluster. Once this is estimated, use the following formula to calculate the total number of blocks required for the cluster:

# blocks = CEIL(# cluster keys / # cluster keys per block)

Note: CEIL means round up.

If you have a test database, you can use statistics generated by the ANALYZE command to determine the number of key values in a cluster key. For example, assume that there are approximately 500 cluster keys in the T1_T2 cluster:

# blocks T1_T2 = CEIL(500/3)  = CEIL(166.7)  = 167

To convert the number of blocks to bytes, multiply the number of blocks by the data block size.

This procedure provides a reasonable estimation of a cluster’s size, but not an exact number of blocks or bytes. Once you have estimated the space for a cluster, you can use this information when specifying the INITIAL storage parameter (the size of the cluster’s initial extent) in your corresponding CREATE CLUSTER statement.

Space Requirements for Clustered Tables in Use

Once clustered tables have been created and are in use, the space required by the tables is usually higher than the estimate derived by the procedure given in the previous section. More space is required because of the method Oracle uses to manage freespace in the database.

See Code Depot for Full Scripts

---

This is an excerpt from Mike Ault, bestselling author of "Oracle 10g Grid and Real Application Clusters". You can buy it direct from the publisher for 30%-off and get instant access to the code depot of Oracle tuning scripts.

Expert Remote DBA BC is America's oldest and largest Remote DBA Oracle support provider.  Get real Remote DBA experts, call BC Remote DBA today.