The most important metrics for measuring the performance of data storage systems are as follows:
Response time, usually reported in milliseconds (ms), is defined as the aggregation of time a request sent to SAN storage, queued, and finally served and responded.
SAB-SE provides a very low response time by optimized storage algorithms and efficient SSD use. One of the important approaches for decreasing response time is RapidStore product that can improve the performance by 2X to 10X.
Throughput refers to the number of transactions per second, known as IOPS that stands for the number of IO operations the storage system can accomplish in one second. IOPS directly depends on the response time. SAB-SE is providing IOPS of 40,000.
One of the features of SAB-SE storage system is compatibility with a big spectrum of workloads. Workloads depend on the type of user data and can be characterized by the request type (read/write), request size, request inter-arrival time, and randomness. We can classify the workloads as follows:
Continuous service, without performance loss, is of major concerns of storage system users. Failure of each storage component such as connection links (Fiber cables, networks…) and disks can result in data unavailability. SAB-SE uses a combination of fault tolerance mechanisms such as RAID and hardware redundancy to assure the highest level of availability.
No single point of failure
Using hardware redundancy we remove almost all single point of failures to assure highest availability.
Using fault tolerance mechanisms, SAB-SE service downtime is less than 5 minutes per year.
SAB-SE is supporting most of high-security platforms as follows:
SAB-SE is designed for scalability in different storage components, removing the need of purchasing new storage systems by increasing the storage needs. This high level of scalability increases the system life-cycle and seriously decreases Total Cost of Ownership (TCO). SAB-SE is scalable in two following aspects:
Of the facilities of SAB-SE data storage system are Instant Copy and Snapshotting.
Despite Instant Copy that saves the copy in the same storage system, Remote Mirroring can store a second copy (mirror) of data in distant storages (the distance can be from a few meters to thousands of kilometers). As the instant copy stores the copy inside the same storage system, it cannot help in events such as unintentional remove of system files, hardware failures, disasters, and terrorist attacks. Remote Mirroring, however, can handle such events and disasters and prevent data loss, as it stores the copy in another distant storage system, using high-speed networks.