lecture 8

Key: a subset of the fields in a record used to identify (uniquely) the record
- Primary key: a key that uniquely identifies a record.
- Secondary key: other keys that may be used for search.

Not every field is a key (keys correspond to fields, or a combination of fields, that may be used in a search)
Four basic types of file organization:
- Sequential
- Indexed
- Indexed sequential.
- Hashed

In all cases, we view a file in a sequence of records. A record is a list of fields. Each field has a data type
File operations:
- Typical Operations: Retrieve a record - Insert a record - Delete a record – Modify a field of a record
- In direct files: Get a record with a given field value
- In sequential files: Get the next record

Sequential files:
- Records are stored contiguously on the storage device.
- read from beginning to end
- Organization of records:
  - Unordered sequential files are called (pile files).
  - Sorted sequential files (records are ordered by some field)

Pile files:
- are a succession of records, simply placed one after another with no additional structure Records may vary in length.
- is the best organization for operations.
- such as: Finding averages, min or max, or sum
- They can be calculated using double buffering as we read though the file once.

To look up a record, given the value of one or more of its fields, we must search the whole file.
In order to search for a record, I have to search for it in the entire file
Exhaustive Reading of the File is Read and process all records (reading order is not important).
When we insert a new record, it will be placed at the end of the file (assuming that we don’t worry about duplicates)

Deleting a record:
- Operations like creating a file, adding records to a file and modifying a record can be performed physically by using basic file operations (open,seek, write, etc.)
- What happens if records are deleted? There is no basic operation that allows us to “remove part of a file”.
- Record deletion should be taken care of by the program responsible for file organization.

Record deletion and Storage compaction:
- Deletion can be done by “marking” a record as deleted (e.g., Place ‘*’ at the beginning of the record
- The space for the record is not released, but the program must include logic that checks if record is deleted or not
- After a lot of records have been deleted, a special program is used to squeeze the file – this is called Storage Compaction.

Deleting fixed-length records and reclaiming space dynamically.
- How to use the space of deleted records for storing records that are added later?
  - Use an AVAIL LIST, a linked list of available records
  - A header record (at the beginning of the file) stores the beginning of the AVAIL LIST.
  - When a record is deleted, it is marked as deleted and inserted into AVAIL LIST.

Deleting variable length records
- Use AVAIL LIST as before but take care of the variable-length difficulties
- The records in AVAIL LIST must store its size as a field. Exact byte offset must be used.
- Addition of records must find a large enough record in AVAIL LIST

There are several placement strategies for selecting a record in AVAIL LIST when adding a new record:
- First-fit Strategy: AVAIL LIST is not sorted by size; first record large enough is used for the new record.
- Best-fit Strategy: List is sorted by size. Smallest record large enough is used
- Worst-fit strategy: List is sorted by decreasing order of size; largest record is used; unused space is placed in AVAIL LIST 21 again.

Rules

b is the total number of blocks in the file.
If the record is at the beginning of the file, then at least 1 block is accessed
If the record is at the end of the file, then at most b blocks are accessed
On average b / 2
(searching) => Time to find and read a record in a pile file is => Time to fetch one record = (b/2) * effective block transfer time
(Reading all records) => Time for exhaustive reading of the file = seek time + rotational delay time + b * time to transfer one block
Time to read the last block = seek time + rotational delay time + block transfer time
Time to insert a new record = Time to read the last block + 2r

Note
(2r - block transfer time): the time to wait for the disk until it rotates back to the original position

Note
(Block transfer time): the time to transfer the new record

Time to update a record for fixed length records = Time to fetch the record + 2r
Time to update a record for variable length records = Time to delete the record + time to insert the new one

`Key:` a subset of the fields in a record used to identify (uniquely) the record

Not every field is a key (keys correspond to fields, or a combination of fields, that may be used in a search)

Four basic types of file organization:

In all cases, we view a file in a sequence of records. A record is a list of fields. Each field has a data type

File operations:

Sequential files:

Pile files:

To look up a record, given the value of one or more of its fields, we must search the whole file.

In order to search for a record, I have to search for it in the entire file

`Exhaustive Reading of the File` is Read and process all records (reading order is not important).

When we insert a new record, it will be placed at the end of the file (assuming that we don’t worry about duplicates)

Deleting a record:

Record deletion and Storage compaction:

Deleting fixed-length records and reclaiming space dynamically.

Deleting variable length records

There are several placement strategies for selecting a record in AVAIL LIST when adding a new record:

Rules

`b` is the `total number of blocks` in the file.

If the record is at the `beginning` of the file, then at `least 1 block` is accessed

If the record is at the `end` of the file, then at most `b` blocks are accessed

On average b / 2

(searching) => Time to `find and read a record` in a pile file is => Time to `fetch` one record = (b/2) * effective block transfer time

(Reading all records) => Time for `exhaustive` reading of the file = seek time + rotational delay time + b * time to transfer one block

Time to `read the last block` = seek time + rotational delay time + block transfer time

Time to `insert a new record` = Time to read the last block + 2r

Time to `update` a record for `fixed length` records = Time to fetch the record + 2r

Time to `update` a record for `variable length` records = Time to delete the record + time to insert the new one

Key: a subset of the fields in a record used to identify (uniquely) the record

Not every field is a key (keys correspond to fields, or a combination of fields, that may be used in a search)

Four basic types of file organization:

In all cases, we view a file in a sequence of records. A record is a list of fields. Each field has a data type

File operations:

Sequential files:

Pile files:

To look up a record, given the value of one or more of its fields, we must search the whole file.

In order to search for a record, I have to search for it in the entire file

Exhaustive Reading of the File is Read and process all records (reading order is not important).

When we insert a new record, it will be placed at the end of the file (assuming that we don’t worry about duplicates)

Deleting a record:

Record deletion and Storage compaction:

Deleting fixed-length records and reclaiming space dynamically.

Deleting variable length records

There are several placement strategies for selecting a record in AVAIL LIST when adding a new record:

Rules

b is the total number of blocks in the file.

If the record is at the beginning of the file, then at least 1 block is accessed

If the record is at the end of the file, then at most b blocks are accessed

On average b / 2

(searching) => Time to find and read a record in a pile file is => Time to fetch one record = (b/2) * effective block transfer time

(Reading all records) => Time for exhaustive reading of the file = seek time + rotational delay time + b * time to transfer one block

Time to read the last block = seek time + rotational delay time + block transfer time

Time to insert a new record = Time to read the last block + 2r

Time to update a record for fixed length records = Time to fetch the record + 2r

Time to update a record for variable length records = Time to delete the record + time to insert the new one

`Key:` a subset of the fields in a record used to identify (uniquely) the record

`Exhaustive Reading of the File` is Read and process all records (reading order is not important).

`b` is the `total number of blocks` in the file.

If the record is at the `beginning` of the file, then at `least 1 block` is accessed

If the record is at the `end` of the file, then at most `b` blocks are accessed

(searching) => Time to `find and read a record` in a pile file is => Time to `fetch` one record = (b/2) * effective block transfer time

(Reading all records) => Time for `exhaustive` reading of the file = seek time + rotational delay time + b * time to transfer one block

Time to `read the last block` = seek time + rotational delay time + block transfer time

Time to `insert a new record` = Time to read the last block + 2r

Time to `update` a record for `fixed length` records = Time to fetch the record + 2r

Time to `update` a record for `variable length` records = Time to delete the record + time to insert the new one