1.4 NSS Features and Capabilities

1.4.1 Use Less Memory and Gain More Speed

NSS requires only about 1 MB of server memory to activate a volume, independent of the number of files it contains. With NSS, you can activate up to 256 NSS volumes concurrently per server, up to the available server memory.

Whenever you activate an NSS volume, it takes only seconds to mount a volume instead of minutes. NSS uses a journaling file system and does not need to scan the entire file system to create a directory entry table (DET) to load the volume. NSS loads a file’s metadata into the memory only when you access the file.

NSS reads the file system journal only if a server goes down abnormally. Instead of slowly searching the volume for errors, NSS reads the journal to identify any incomplete transactions. It either completes the transaction or backs it out. This results in less server down time and is beneficial for applications such as mail services.

1.4.2 Improve Storage Availability

NSS provides the following features to improve I/O performance and provide fault-tolerant access to your data:

1.4.3 Prevent Unauthorized Access

NSS includes the following features to help prevent access to data that circumvents normal access control:

1.4.4 Protect Data from Corruption or Loss

NSS includes the following features to ensure that the most current copy of your data is recoverable:

1.4.5 Maximize Available Space

NSS includes the following features to help you maximize your available space:

1.4.6 Delayed Block Allocation

Delayed block allocation in NSS improves the locality of sequential file content on a disk and thereby reduces the disk fragmentation.

Buffered writes is the only supported write mechanism in NSS. During buffered writes, delayed block allocation allows aggregation of sequential file blocks before writing them to the disk. The aggregation of sequential file blocks allows all the blocks to be allocated as a single extent (set of contiguous disk blocks) instead of separate disk blocks, if they were allocated at user write time. This locality of sequential file blocks on disk helps to improve the access time during the sequential read work loads such as filesystem reads or backup.

Even writes of such blocks into disks perform better as it minimizes the rotational and seek latencies involved in the movement of disk head in a rotational disk, compared to the traditional block allocation.

Delayed block allocation also reduces the writes of NSS metadata such as journal, fileMap, freeTree and logged pool or volume as it now updates each of these metadata for all aggregated sequential file blocks instead of individual user file block writes.

This feature is not applicable for the volumes that enabled with Compression, User quota, and Directory quota features.