At our recent SNIA Cloud Storage webcast â€œIP-Based Object Drives Now Have a Management Standard,â€ our panel of experts discussed how the SNIA release of the IP-Based Drive Management Standard eases the management of these drives. If you missed the webcast, you can watch it on-demand.
A lot of interesting questions came up during the live event. As promised, here are answers to them:
Q. Am I correct in thinking that each IP based drive will have a unique IP address?
A. Each Ethernet interface on the drive will have its own unique IP Address. Object Drives may be deployed in private address spaces (such as in a fully configured rack). In such configurations, two Object Drives might have the same IP address, but would be on completely separate networks.
Q. Assuming vendors will be using RedFish, will the API calls be made through existing middleware or directly to the BMCs (baseboard management controllers, specialized service processors that monitors the physical state of a computer) on the platforms?
A. Redfish can be supported by host based middleware, the enclosure’s BMC, or may be supported directly from the drive.
Q. Would a drive with native iSCSI protocol and an Ethernet interface be considered an “IP Drive”?Â Â
A. Yes. This is why we use the generic IP Drive term as it allows for multiple protocols to be supported.
Q. What are the data protection schemes supported in the existing products in this space?
A. Examples of data protection typically used with IP drives include erasure encoding and traditional RAID.
Q. Is this approach similar to the WD Ethernet Drive?
A. The WD Ethernet Drive is an IP based drive.
Q. Do you expect to see interposers with higher Ethernet bandwidth that could be used with SSD vs. HDDs?
A. Yes, there are multiple examples starting to appear in the market of interposers for SSDs.
Q. Is this regular Ethernet or NVMe over Fabrics?
A. Regular Ethernet. This does not require Converged Ethernet, nor anything layered on that. NVMe over Fabrics could utilize IP based Drive Management in the future.