Seagate's EVault cloud backup sub has launched a Glacier cloud archive competitor, priced at $15 per TB per month with instant data access and data preserved intact for decades. It will use Seagate Kinetic drives in the future.
Amazon started the cloud archive ball rolling big time with its Glacier service costing $0.01/GB/month with data retrieval access taking three - five hours. Retrieval can also cost money but Amazon says customers get five per cent of retrievals free each month.
LTS2 is EVault's Long Term Storage Service, with data stored in EVault data centres using Seagate disk drives. The service effectively costs $0.015/GB/month; you pay half as much again as Amazon for the much faster retrieval. It is an object-based storage scheme, built on the OpenStack Swift technology, and features:
- Comprehensive SLAs for availability and data durability, portability, and survivorship.
- Continuous data integrity checking and self-healing.
- Multiple copies of data.
- Scalable—allows unlimited number of objects and containers.
- Physical media for bulk data import and export.
- Instant retrieval as it's disk-based - < five seconds first-byte latency.
- Access via Swift API, S3 API, existing backup/archive applications and 3rd-party cloud storage gateways.
- Metadata services for categorising and finding data.
- Auditable event tracking for data access monitoring.
- SSAE 16-compliant data centre infrastructure in Bluffdale, Utah.
- In-flight and at-rest data encryption via cloud gateway.
The less than five seconds to first byte latency suggests that spun-down disks could be being used, although EVault isn't supplying any details in this area. We also suspect Seagate's shingled magnetic recording (SMR) drives are being used but EVault is staying schtum about that too.
Seagate says it's shipped a million SMR drives but they have not been publicly announced and Facebook, a known SMR drive using customer, is trialling WD SMT spindles, not Seagate ones. So, maybe, EVault, Seagate's subsidiary, is the mystery recipient of a million SMR Seagate drives.
More EVault LTS2 data centres with geo-replication will be opened world-wide in 2014 to better comply with local data protection and privacy laws.
EVault has been engaged in an OpenStack project and the LTS2 service is a direct result of that. It says it's a trustworthy supplier because it's backed by Seagate and has some 43,000 companies using its existing cloud backup service. It says LTS2 has up to 99.999999999 per cent data durability over a given year - or "eleven nines". It claims it is "the only cloud storage provider to include a Service Level Agreement (SLA) for durability."
You can transfer all or a portion of your data without transfer or request fees as long as the transfer amount of data does not exceed your average monthly storage volume over the past 2 years. For data transfer out from EVault LTS2 to Internet it's $0.07/GB over 5 per cent of the average amount of data stored.
EVault says it's "targeting any long-term storage use case where the read is infrequent and the content stored is rich-data (i.e. large objects). Some examples are media and entertainment, medical imaging, magnetic tape replacement, and the protection of cultural and corporate history."
You can read more about LTS2 and EVault's OpenStack project in a blog written by EVault's Mikey Butler, VP for IaaS research and development. He writes that: "Over the next five years EVault and Seagate will create the world’s largest, most durable, cost effective, easiest to adopt, disk archival cloud."
Butler gives an insight into the scale of the storage involved: "Our archive cloud must exceed 8 exabytes to achieve our pricing objectives. That’s 2 million 4TB drives without resiliency overhead! ... Our target is 30-40 per cent resiliency overhead at most." Which means 2.8 million spindles at 40 per cent.
The rate of disk faults at the scale is stupendous: "Modeling done at a major university for Sandia Labs shows that at 1 million drives one can expect a significant disk fault every 6 seconds." That rules RAID and its associated rebuilds out and makes object storage with its self-healing characteristics the only feasible way to go.
The current eleven-nines of data durability isn't enough: "We want 13+-nines of durability with objects distributed across multiple disks, storage nodes, data centres, [and] geographic risk zones."
The blog conforms that spin-down disks are used: "We want our cloud to have 93 per cent of the drives powered down at any one time with the remaining 7 per cent powered up and providing object location and health information."
Kinetic drives coming
Butler says LTS2 will use Seagate's Kinetic drives: "Kinetic is a dramatic re-imagining of the disk drive. Instead the familiar SATA, SAS or SCSI block storage device, Kinetic is an Ethernet-connected, object-friendly key value drive with a generous amount of compute on board! Now disk drives can do much more on their own helping us more easily address many of the above mentioned challenges. For example, with Kinetic, it is now possible for drives to do object consistency checking and inter-drive object migration on their own without servers being in the data path, which implies fewer servers ... LOVE IT!"
Bootnote See also Quantum's Q-Cloud as an alternate cloud backup and archive service, priced from as little as $0.01/GB/month once you have 72TB of stored data, and faster data retrieval including local storage-based restores off a front-end DXi appliance.