The CAStor FSG provides CAS functionality through standard file interfaces at price points that was not achievable before the company’s introduction of CAStor software, according to Caringo of Austin, Texas. CAStor FSG implements CAStor quickly without requiring any code, making any changes to existing applications, and without any complex integration, Caringo said.
CAStor FSG runs on a server that is built around standard x86-based commodity hardware platform, and connects to the CAStor Cluster through standard Gigabit Ethernet (GbE), the company added. CAStor FSG is fully multi-threaded and uses connection pooling to maximize throughput to and from the CAStor Cluster, it added. Additional CAStor FSG nodes can be added to provide more access points into CAStor to allow parallel scalability, Caringo said.
The benefits of CAS are best realized in vertical segments where there are industry-specific business applications that generate the bulk of content, like medical imaging, security camera footage, newspaper archives, voice and email messages and check image storage, Caringo said. Many such applications require legacy file systems using local disks, NAS (network attached storage) or SAN (storage area network), it added. With CAStor FSG, the company is enabling application providers to deploy and integrate CAStor easily, creating a lower overall cost of ownership, freedom from vendor-proprietary CAS hardware, and increased margins for partners, Caringo added.
CAStor FSG allows administrations to set policies, permissions, and constraints for fixed content to meet internal goals and also to assure regulatory compliance, Caringo said. Once these policies are configured, CAStor transparently monitors data and enforces the policies to ensure the correct number of replicas for a particular file, verifying file lifecycle information and managing secure deletion, it added. A multi-threaded, client pooling architecture provides throughput and scalability, Caringo added.
CAStor FSG allows any application that works with a CIFS/NFS file system to connect to CAStor, and provide various ‘views’ into the content store, Caringo said. Classic file system mapping provides a virtual file system front end to CAStor that users can access through standard network file systems and protocols, it added.
Unix/Linux users can mount CAStor FSG as a network drive through NFS, according to Caringo. Microsoft Windows users can map a Windows network drive through CIFS, where the CAStor Cluster would look like a Windows file system, it added. Mac OS X users can use both approaches, the company added.
Once the fixed content and reference information has been stored in CAStor, it can be accessed through Simple Content Storage Protocol (SCSP), CAStor’s HTTP 1.1 subset, using the Universally Unique Identifier (UUID)’s stored in the database, or as files in a regular file system, it added. This allows for phased migration strategies to a native CAStor interface, Caringo added.
The CAStor FSG also features a standard MySQL database which is already configured and ready to use, offers full support for Oracle databases, and provides high availability for databases and other critical applications through mirroring, Caringo said. The file system also offers database querying to discover information like file name, UUID mapped to the file, last access, last modification, permissions, file size and type and user ID, the company added. The CAStor FSG is capable of standalone operation or integration with existing user directories using standard authentication methods such as Microsoft’s Active Directory or Kerberos, it added. The file system also allows straightforward administration and reporting capabilities, Caringo added.
The CAStor FSG is available as a Beta version from this month, the company added.
