Overview
This section describes the background, definition, and benefits of SmartPartition.
Background
An IT system architecture typically consists of three parts: computer, network, and storage. In traditional system architecture, application systems are independent from each other, and a single storage system supports only a small number of applications that have a minor performance impact on each other.
With constant development and improvement, a storage system nowadays has to provide storage for thousands of applications. Due to such a sharp application growth and increasingly varied application I/O characteristics, applications contend with each other for resources. As a result, the performance of critical applications cannot be ensured.
To deliver high QoS, storage vendors have come up with a number of quality assurance technologies, such as I/O priority setting, application traffic restriction, and cache partitioning. Because cache resources are a necessary path for data to flow between storage systems and front-end application systems, cache partitioning provides an important way to improve QoS.
Definition
SmartPartition is an intelligent cache partitioning feature developed by Huawei. It allows you to assign cache partitions of different sizes to different applications. The system provides the cache capacity specified by a partition for the corresponding applications, thereby ensuring the performance of the applications in the partition.
SmartPartition is supported for LUNs (block services) and file systems (file services).
Benefits
Table 1-1 describes the benefits of SmartPartition.
Benefits |
Description |
---|---|
Separated resources for services |
Applications use their own cache partitions, preventing a mutual impact between applications of different types and ensuring the overall QoS. |
Ensuring cache resources for critical applications |
Cache resources provided for non-mission-critical applications are restricted so that more cache resources are available to mission-critical applications, ensuring the performance of mission-critical applications. |