Available RAM that NetWare uses to improve NetWare server access time.

Cache memory allocates memory for the hash table, the FAT, the Turbo FAT, suballocation tables, the directory cache, a temporary data storage area for files and NLM files, and available memory for other functions.

Directory Caching

A method of decreasing the time it takes to determine a file's location on a disk.

The FAT and DET are written into the server's memory. The area holding directory entries is called the directory cache.

The server can find a file's address from the directory cache much faster than retrieving the information from disk.

Figure 4
Directory Caching

A server can service requests from workstations up to 100 times faster when it reads from and writes to the server's cache memory, rather than reading from or writing to the server's hard disks.

The following figure illustrates how a workstation makes a read request from the server and how the server executes a hash algorithm to calculate the file's address from a hash table.

Figure 5
File Caching

Hashing is a quick way of predicting the file's address in the directory table.

For example, the address on the hash table contains pointers to the first and second probable locations of the DB.DAT file's directory entry in the directory cache.

If the first search isn't successful, the server uses the second pointer to find the directory entry, as illustrated in the following figure:

Figure 6
Locating the Directory Entry

When the directory entry is located, the server first checks its cache memory to see if it has a copy of DB.DAT.

Figure 7
Locating the File in the Cache Memory

If the file is there, the server sends the file to the station from memory. If the file isn't there, the server retrieves the file from disk and sends it to the station.

Writing Files to Cache

When a workstation writes a file to the server, the server performs the hash algorithm to find the file's cache buffer.

It writes the file to the designated location (writing over the old file, if the file was already in cache) and updates the directory table in the directory cache.

Since the file has changed, its cache buffer is different from the file on disk. The buffer is then referred to as a dirty cache buffer.

Since writes to disk take longer to perform than writes to cache, the server keeps the dirty buffer designation on the file in cache until the disk has received the file.

Writing a file to cache is illustrated in the following figure:

Figure 8
Writing a File to Cache

The server then sends a message to the station that the server has received the file, and the station is free to complete other processes.

Once the file is written to disk, the server checks the data in memory against the data on disk. If there is a match, the buffer is no longer dirty, as illustrated in the following figure:

Figure 9
Cache Memory Write and Verify

As the cache memory fills up, buffers containing the least-used files and directories are eliminated.