Before reviewing performance data, review the key terminology used to describe storage performance as these terms form the base for understanding performance statistics.
These terms are used frequently to describe storage performance:
- Input/Output per second
- Measurement of the number of input and output operations completed per second. Input/Output per second (IOPS) is usually shown as read, write, and total IOPS.
- Latency
- When reading and writing data, latency is defined as the time it takes for a read operation or write operation to complete. Factors that contribute to latency include the amount of time to read data and write data from a disk, and the amount of time required to transfer the data over a storage link.
- Queue Depth
- Queue depth, sometimes referred to as I/O queue depth, is the number of I/O requests (pending or outstanding I/Os) waiting to be completed.
- Response Time
- Response time is the average time, in milliseconds, to complete an I/O operation.
- Throughput
- Throughput is measured as the amount of data, in megabtyes per second, that is processed by the Oracle FS System. Throughput is also expressed in IOPS when measuring sequential data. Whether expressed as megabtyes per second or as IOPS, throughput represents bandwidth.