Bereitgestellt von: IBM Deutschland GmbH
This is a technical whitepaper on the state-of-the art in high speed transport direct-to-cloud storage and support for third party cloud storage platforms.
The majority of cloud-based storage available in the market place today is based on object storage. Key design principles of object storage architectures are the separation of file data and metadata, replication of data across distributed commodity storage, and unified access across distributed nodes and clusters. These principles enable more cost-effective scale-out with greater redundancy and durability than traditional blockbased storage.
However, these same attributes create challenges for storing large unstructured data. Large files must be divided into “chunks” and stored independently on “writes”, and reassembled on “reads”. When coupled with the traditional
bottleneck of moving large files over long-haul WAN connections with high round-trip time and packet loss, cloud-based object storage becomes impractical for large
unstructured data due to the dramatic reduction in transfer speed and throughput, and extended delays in transferring and storing the files.
Bei Problemen mit dem Öffnen des Whitepapers deaktivieren Sie bitte den Popup-Blocker für diese Website. Sollten Sie weitere Hilfe benötigen, verwenden Sie bitte unser Kontaktformular.
Publiziert: 01.07.15 | IBM Deutschland GmbH