One, background
In recent years, faced with the challenges of digital transition, more and more companies began to cloud over business-critical systems, there are more business innovation in the cloud to help companies achieve business growth, these data has become the most important business assets, resources. For businesses, how to ensure valuable data is not lost assets, because of data does not affect normal access to business data has been an important challenge to face.
Ali cloud object storage publish city redundant storage types, providing data center-class (available district-level) disaster recovery capabilities when off network, power failure or catastrophic event such as extreme events leading to a room (usable area) is not available, still be able to ensure continued strong consistency of service capabilities to meet business-critical systems for the "recovery time objective (RTO)" and "recovery point objective (RPO)" strong demand is equal to zero.
Second, the principle of analytical techniques
Aliyun OSS city redundant memory, as the name implies, the redundant data is distributed across the available area of the city, which is used in areas distributed across the available technology erasure codes, erasure codes conventional technology, generally deployed in the same cluster, you can deal with such failed host level, with respect to the multi-copy technology, to better balance its space utilization and data redundancy to achieve.
OSS city erasure code using redundant memory 6 + 6, 12 to break the data distributed block redundancy is available on 3 areas, there are four blocks of data available for each area, the maximum erasure codes mechanism 6 + 6 can allow damaged or lost 6 blocks. So when one of the available area (or room) encountered extreme cases cause the entire available area can not serve, will only affect four blocks, will not affect the reliability of business continuity and data. Throughout the entire failover process without user awareness, services are not interrupted, data is not lost.
At the same time Ali cloud OSS city redundant storage using a strong consistency model to ensure that exactly the same data, no stale read. In addition, to ensure that the overall literacy, providing usable area before the same level Tbps bandwidth, let the data flow rapid flow, ensure sufficient distance between each of the available area, city rolling minimize disaster risks. Finally, the OSS city aliyun redundant memory can be provided to a user 99.9999999999 percent (12 9) data reliability and data availability SLA 99.95%, and when the total failure of the engine room, to achieve RPO = 0, RTO = 0.
Third, the best practice - "two to three centers" architecture
The traditional offline data center, many industry regulatory requirements demanding enterprise core system must meet the "three centers in two" framework to ensure foolproof core assets. Redundant storage by the city combined with the existing cross-boundary copy function, you can easily build "three centers in two" architecture.
At this point, Ali cloud OSS can provide business users with a different room, city, cross-regional full service multi-level disaster recovery capabilities.
View: object storage OSS
a map to quickly understand: release core, scene, benefits, access, more!
Live Review: OSS conference across the city 3AZ
Click for: "Ali cloud the new conference channel"
Subscribe Now: Ali cloud the new conference · Weekly