Cartika ZFS Storage Solution for Cloud Computing

Company Info - Technical Documents

In computing, ZFS is a combined file system and logical volume manager designed by Sun Microsystems. The features of ZFS include data integrity verification against data corruption modes (like bit rot), support for high storage capacities, integration of the concepts of filesystem and volume management, snapshots and copy-on-write clones, continuous integrity checking and automatic repair, RAID-Z and native NFSv4 ACLs. ZFS is implemented as open-source software, licensed under the Common Development and Distribution License (CDDL). The ZFS name is a trademark of Oracle. To read more about ZFS filesystems and logical volume managers, please click HERE.

Cartika has developed a cloud storage platform to operate within the Oracle Open Standards and Open Source ZFS filesystems and logical volume managers. Using this technology, Cartika is able to deliver highly flexible and available storage systems which are resilient to corruption by perpetually self healing itself. The Cartika ZFS storage solution is infinitely scalable and offers the highest possible performance available in the industry by utilizing extensive, flexible and scalable SSD caching for BOTH reads and writes. 

Our ZFS storage solution also offers the unique advantage of being completely flexible via the creation of zpools across virtual storage pools. Virtual storage pools can be comprised of entire shelves of disks or a combination of disks from various sources. zpools can be redundant over multiple devices and over multiple locations. Cartika configures zpools to be redundant within individual shelves, as well as across multiple shelves. We maintain the ability to lose a complete shelf (as well as a complete head unit) without impacting service and without any data loss. We are able to scale single zpools to multiple petabytes worth of usable drive space - seamlessly and affordably.

Cartika automatically replicates all cloud data between the United States and Canada for the purpose of disaster recovery at no additional cost.  All Cloud server customers with data on our ZFS clusters will have a local replication of their data on completely different SANs and the data will be automatically replicated off site and to a different country daily for the purpose of disaster recovery if ever required.  For customers who do not wish their data to be replicated to a different country simply need to make that request and it will be excluded from the off site replication - but, will maintain the daily replication to our alternate ZFS backup SAN's.

From a performance perspective, the most significant advantage of the Cartika ZFS storage solution is the SSD caching outlined below.

Cartika SSD Read & Write Cache SSD caching enables pure SSD level performance for the SAN's disk arrays All data being read or written to or from the disk arrays are handled by the SSD caching layer (500 Mbps Reads and Writes) All SSD read and write caches are completely redundant offering zero data loss in case of failure to any individual SSD cache drive SAN's can never be hit with IO bottlenecks associated with drive speeds or # of drive spindles that traditional SANs are limited by as all data is served and buffered by the SSD caching layer In case the SSD caching layer becomes saturated, the Cartika ZFS system is able to seamlessly grow its volume using available SSD drives anywhere within the ZFS Cluster. This ensures all customers reads and writes are performed at continual and perpetual SSD speeds vs SATA or SAS speeds The ZFS cluster is also able to take available RAM from the system head units to act as caching devices if required.