|
 |
|
One of the unique features of OpenSolaris is the ZFS file system. Most likely, one of the main reasons you are interested in OpenSolaris is because of the fame that this powerful, yet easy-to-use, file system has gained.
|
|
The motivation behind ZFS is to make available to everyone the flexibility of the pooled storage model that large-scale storage systems provide, but without the complex administration and high cost of those storage systems. By integrating management of the disks and the file systems together, ZFS tries to make your storage as self-managing as your system's memory. ZFS is designed to scale to extremely large quantities of data by using 128-bit data addressing and dynamically scaling its metadata, rather than using the fixed scales demanded by UFS and other file systems of its generation, which weren't designed for terabyte and largar scales.
|
|
ZFS provides high performance through a fully parallel design with an I/O pipeline that's modeled on the concepts of CPU instruction pipelines. By using a transactional, copy-on-write update model, ZFS ensures that its data is always consistent on disk. ZFS computes a checksum on every data and metadata block and because its check sums cover the entire data path and are stored separately from the data being checksummed, it can detect data corruption caused by any element of the storage subsystem, not just disk errors. As a result, there is no need for a traditional file system checking and repair utility such as fsck, and inexpensive disks can provide similar reliability to high-priced storage systems.
|
|
|
 |
|
One of the unique features of OpenSolaris is the ZFS file system. Most likely, one of the main reasons you are interested in OpenSolaris is because of the fame that this powerful, yet easy-to-use, file system has gained.
|
|
The motivation behind ZFS is to make available to everyone the flexibility of the pooled storage model that large-scale storage systems provide, but without the complex administration and high cost of those storage systems. By integrating management of the disks and the file systems together, ZFS tries to make your storage as self-managing as your system's memory. ZFS is designed to scale to extremely large quantities of data by using 128-bit data addressing and dynamically scaling its metadata, rather than using the fixed scales demanded by UFS and other file systems of its generation, which weren't designed for terabyte and largar scales.
|
|
ZFS provides high performance through a fully parallel design with an I/O pipeline that's modeled on the concepts of CPU instruction pipelines. By using a transactional, copy-on-write update model, ZFS ensures that its data is always consistent on disk. ZFS computes a checksum on every data and metadata block and because its check sums cover the entire data path and are stored separately from the data being checksummed, it can detect data corruption caused by any element of the storage subsystem, not just disk errors. As a result, there is no need for a traditional file system checking and repair utility such as fsck, and inexpensive disks can provide similar reliability to high-priced storage systems.
|
|
|
|
