ZFS: The File System of the Future

Slide 1: ZFS Click to edit the title text format THE FILE SYSTEM OF THE FUTURE Danilo Poccia ● Presenter’s Name –Presenter’s Title Senior Systems Engineer ●Presenter’s Company Sun Microsystems Italia S.p.A.

Slide 2: What is ZFS? A new way to manage data End-to End Immense Data Software Data Integrity Capacity Developer With check- The world's summing and first 128-bit copy-on-write file system transactions Easier Huge Performance Administration Gains A pooled Especially storage model – architected no volume for speed manager

Slide 3: No More Volume Manager! Automatically add capacity to shared storage pool Application 1 Application 2 Application 3 ZFS ZFS Storage Pool

Slide 4: Easier Administration • Pooled Storage Design makes for Easier Administration No need for a Volume Manager! • Straightforward Commands and a GUI > Snapshots & Clones > Quotas & Reservations > Compression > Pool Migration > ACLs for Security

Slide 5: Quotas and Reservations • To control pooled storage usage, administrators can set a quota on a per file system basis > Limit Tim to a quota of 10g # zfs set quota=10g tank/home/tim • Or they can set a reservation (minimum) > Guarantee Fred a reservation of 20g # zfs set reservation=20g tank/home/fred

Slide 6: Quotas and Reservations • To control pooled storage usage, administrators can set a quota on a per file system basis • Or they can set a reservation (minimum) ZFS > Limit Tim to > Guarantee a quota of Fred a 10g reservation # zfs set of 20g quota=10g tank/home/tim # zfs set reservation=20g tank/home/fred Storage Pool Usage

Slide 7: Copy-on-Write and Transactional Uber-block Original Data New Data Initial block tree Writes a copy of some changes Original Pointers New Uber-block New Pointers Copy-on-write of indirect blocks Rewrites the Uber-block

Slide 8: End-to-End Checksums Checksums are separated from the data Prevents: > Silent data corruption > Panics from corrupted metadata > Phantom writes > Misdirected reads and writes > DMA parity errors > Errors from driver bugs > Accidental overwrites Entire I/O path is self-validating (uber-block)

Slide 9: RAID-Z Protection RAID-5 and More • ZFS provides better than RAID-5 availability • Striping uses dynamic widths > Each logical block is its own stripe • All writes are full-stripe writes > Eliminates read-modify-write (So it's fast!) • Eliminates RAID-5 “write hole” > No need for NVRAM

Slide 10: Self-Healing Data ZFS can detect bad data using checksums and “heal” the data using its mirrored copy. Application Application Application ZFS Mirror ZFS Mirror ZFS Mirror Detects Bad Data Gets Good Data from Mirror “Heals” Bad Copy

Slide 11: ZFS Snapshots • Provide a read-only point-in-time copy of file system • Copy-on-write makes them essentially “free” • Very space efficient – only changes are tracked • And instantaneous – just doesn't delete the copy Snapshot Uber-block New Uber-block Current Data

Slide 12: ZFS Clones Writable copy of a snapshot Ideal for storing many private copies of shared data: ● Software installations ● Workspaces ● Diskless clients Example: Create a clone of your OpenSolaris source code # zfs clone tank/solaris@monday tank/ws/lori/fix

Slide 13: Data Compression • Reduces the amount of disk space used • Reduces the amount of data transferred to disk – increasing data throughput ZFS Data Compression

Slide 14: “Adaptive Endian-ness” - Hosts always write in their native “endian-ness” Opposite “Endian” Systems - Write and copy operations will eventually byte swap all data! Config Data is Stored within the Data - When the data moves, so does its config info Storage Pool Migration

Slide 15: Copy-on-Write Design Multiple Block Sizes Pipelined I/O Dynamic Striping Intelligent Pre-Fetch Architected for Speed

Slide 16: ZFS edit the title text format Click to THE FILE SYSTEM OF THE FUTURE Danilo Poccia ● Presenter’s Name –Presenter’s Title danilo.poccia@sun.com ●Presenter’s Company blogs.sun.com/danilop 16