Got problems with storage capacity? Sun may have just the solution you've been waiting for: the SolarisTM Zettabyte File System (ZFS), a radically new approach to data management.

What distinguishes Solaris ZFS from other storage solutions is that it eliminates the volumes, partitions, provisioning, wasted bandwidth and stranded storage associated with its predecessors. Instead Solaris ZFS employs a common storage pool from which every one of thousands of file systems can draw at all times. Furthermore, Solaris ZFS does not overwrite data but rather saves the new data first and then deletes the information that it replaces, preventing data loss in case of a system outage. Built-in checks prevent data corruption.

As the world's first 128-bit file system, Solaris ZFS has 18 billion times the storage capacity of current generation 64-bit systems. There are no arbitrary limits in Solaris ZFS, which can have as many files as necessary: full 64-bit file offsets; unlimited links, directory entries, snapshots, and so on. It has been released under the same terms as OpenSolaris, which means the open source Common Distribution and Development License (CDDL) applies.

Because all operations are copy-on write transactions, the on-disk state is always valid, meaning there is never a need to fsck(1M) a Solaris ZFS. Checksums of each block prevent silent data corruption. Data is self-healing in replicated (whether mirrored or RAID) configurations. Solaris ZFS will detect damaged copies and replace them with another copy.

The new data replication model, RAID-Z, featured in Solaris ZFS eliminates the old RAID-F write hole by creating full-stripe writes, thereby eliminating the need for NVRAM in hardware as well and making it possible to use cheap media. A disk scrubbing feature protects against disk failure.

The pipeline I/O engine in Solaris ZFS uses I/O dependency graphs that provide scoreboarding, priority, deadline scheduling, out-of-order issue and I/O aggregation, all of which gives Solaris ZFS a performance level that far surpasses that of other filesystems.

Solaris ZFS also features unlimited constant-time snapshots and clones that provide read-only pont-in-time copies of a filesystem and a writable copy of a snapshot, respectively. Snapshots power Solaris ZFS backup and restore that includes both full and incremental backup. The incremental backups are so efficient that they can be used for remote replication, transmitting an incremental update every 10 seconds.

The build-in compression featured in Solaris ZFS reduces space usage by two to three times and reduces the amount of I/O by a similar factor, with the result that workloads go faster. The volumes that Solaris ZFS storage pools can provide when raw-device semantics are called for can be used as swap devices, which on a swap volume, results in compressed virtual memory. [...read more...]

fullsource

From the latest issue:

Recent Blog Entries as of March 22, 2010, 2:56 am
Oracle Works to Assuage the Anxieties of Its Customers SysAdmin News Bites Oracle Enterprise Manager Ops Center Released Suits and Their Countersuits A More Direct Channel Model Planned for Oracle-Sun	The Importance of Information Storage and Management Sun POJO Service Engine User's Guide Sun Storage 7000 2010.Q1 Quality and Features PSA Deploys GlassFish in the Manufacture of Peugeot, Citroën Vehicles Running VirtualBox VMs as Services in Windows
Blog RSS Feed