Checklist for Storage Performance & Availability
When it comes to Storage Performance and Availability, we always try to compare different storage vendors and systems to analyze the parameters.
Here I have tried to list down all the common parameters which are applicable to all storage vendors and can be used as a general baseline for understanding and troubleshooting the performance related issues.
- Availability – is related the hardware components of the storage system and to avoid single point to failure.
- Bandwidth – is the measurement of amount of data that can be transferred along a channel per second. Some storage systems may have multiple ports and we need to identify if they are configured as active/standby or load balance.
- Response Time – the time taken for a request of service and the completion of request.
- RPO – Recovery Point Objective (RPO) is the amount of data at risk or data loss an organization can afford in event of Disaster / Failure.
- RTO – Recovery Time Objective (RTO) is the time required to bring an application or service back online state after a Disaster / Failure.
- I/O Characteristics – it is depended on the type of workload or application framework such as OLTP, Email, File serving, Backup to Disk or Rich Media
- Knowing the workload – to implement best practice to achieve maximum performance from your storage system, it is important to understand the workloads that are going to store data on the storage. This requires a basic knowledge of the host applications.
- Throughput – Also known as IOPS. This is the number of I/O operations that are processed per second. It can be divided into Read IOPS, Write IOPS & Total IOPS