Performance metrics for storage system controllers on DS8000 storage systems

You can include front end data, cache data, back-end data, and other data for storage system controllers on DS8000 series systems in performance reports.

Front end data

You can create performance reports that include the following information:

Normal Read I/O Rate (ops/s): The average number of nonsequential read operations per second.
Sequential Read I/O Rate (ops/s): The average number of sequential read operations per second.
Overall Read I/O Rate (ops/s): The average number of read operations per second. This value includes both sequential and nonsequential read operations.
Normal Write I/O Rate (ops/s): The average number of nonsequential write operations per second.
Sequential Write I/O Rate (ops/s): The average number of sequential write operations per second.
Overall Write I/O Rate (ops/s): The average number of write operations per second. This value includes both sequential and nonsequential write operations.
Total Normal I/O Rate (ops/s): The average number of nonsequential I/O operations per second. This value includes both read and write operations.
Total Sequential I/O Rate (ops/s): The average number of sequential I/O operations per second. This value includes both read and write operations.
Total Overall I/O Rate (ops/s): The average number of read operations and write operations per second. This value includes both sequential and nonsequential operations.
Record Mode Read I/O Rate (ops/s): The average number of I/O operations per second for record-mode read operations.; For record-mode read operations, only the requested data is managed in the cache rather than a full track of data.
Read Data Rate (MiB/s): The average number of MiBs per second that are transferred for read operations.
Write Data Rate (MiB/s): The average number of MiBs per second that are transferred for write operations.
Total Data Rate (MiB/s): The average number of MiB per second that are transferred for read operations and write operations.
Read Response Time (ms/op): The average number of milliseconds to complete a read operation.
Write Response Time (ms/op): The average number of milliseconds to complete a write operation.
Overall Response Time (ms/op): The average number of milliseconds to complete an I/O operation. This value includes both read and write operations.
Read Transfer Size (KiB/op): The average number of KiB that are transferred per read operation.
Write Transfer Size (KiB/op): The average number of KiB that are transferred per write operation.
Overall Transfer Size (KiB/op): The average number of KiB that are transferred per I/O operation. This value includes both read and write operations.

Cache data

You can create performance reports that include the following information:

Write Cache Delay I/O Rate (ops/s): The average number of I/O operations per second that are delayed because of space constraints in the write cache, or because of other conditions.; This value applies only to resources that are running a version of IBM Storage Virtualize earlier than 7.3. The metric is an approximation because actual transfer rates are different for each cache layer. Therefore, you cannot directly compare this metric from resources that are 7.3 or later with earlier versions.
Normal Read Cache Hit Percentage: The percentage of nonsequential read operations that find data in the cache.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because a cache miss requires access to the back-end storage resources.
Sequential Read Cache Hit Percentage: The percentage of sequential read operations that find data in the cache.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because a cache miss requires access to the back-end storage resources.
Overall Read Cache Hit Percentage: The percentage of all read operations that find data in the cache. This value includes both sequential and random read operations, and read operations in the volume cache and volume copy cache where applicable.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because in the event of a cache miss, the data must be read from the back-end storage resources.
Normal Write Cache Hit Percentage: The percentage of nonsequential write operations that are handled in the cache.
Sequential Write Cache Hit Percentage: The percentage of sequential write operations that are handled in the cache.
Overall Write Cache Hit Percentage: The percentage of all write operations that are handled in the cache. This value includes both sequential and random write operations, and write operations in the volume cache and volume copy cache where applicable.
Total Normal Cache Hit Percentage: The percentage of nonsequential I/O operations that are handled in the cache. This value includes both read and write operations.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because a cache miss requires access to the back-end storage resources.
Total Sequential Cache Hit Percentage: The percentage of sequential I/O operations that are handled in the cache. This value includes both read and write operations.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because a cache miss requires access to the back-end storage resources.
Total Overall Cache Hit Percentage: The percentage of all read operations and write operations that are handled in the cache. This value includes both sequential and random read and write operations, and read and write operations in the volume cache and volume copy cache where applicable.; You can use this value to understand throughput or response times. Low cache-hit percentages can increase response times because a cache miss requires access to the back-end storage resources.
Record Mode Read Cache Hit Percentage: The percentage of cache hits for record-mode read operations.; For record-mode read operations, only the requested data, rather than a full track of data, is managed in the cache.
Write Cache Delay Percentage: The percentage of I/O operations that are delayed because of space constraints in the write cache, or because of other conditions. The value is a percentage of all operations.; This value applies only to resources that are running a version of IBM Storage Virtualize earlier than 7.3. The metric is an approximation because actual transfer rates are different for each cache layer. Therefore, you cannot directly compare this metric from resources that are 7.3 or later with earlier versions.
Disk to Cache Transfer Rate (ops/s): The average number of sectors or tracks per second that are transferred per second from the disks to the cache. This value applies only to resources that are running a version of IBM Storage Virtualize earlier than 7.3. The metric is an approximation because actual transfer rates are different for each cache layer. Therefore, you cannot directly compare this metric from resources that are 7.3 or later with earlier versions.
Cache to Disk Transfer Rate (ops/s): The average number of sectors or tracks per second that are transferred from the cache to the disks. This value applies only to resources that are running a version of IBM Storage Virtualize earlier than 7.3. The metric is an approximation because actual transfer rates are different for each cache layer. Therefore, you cannot directly compare this metric from resources that are 7.3 or later with earlier versions.

Back-end data

You can create performance reports that include the following information:

Back-End Read I/O Rate (ops/s): The average number of read operations per second that are issued to the back-end storage resources.
Back-End Write I/O Rate (ops/s): The average number of write operations per second that are issued to the back-end storage resources.
Total Back-End I/O Rate (ops/s): The average number of I/O operations per second that are transmitted between the back-end storage resources and the component. This value includes both read and write operations.
Back-End Read Data Rate (MiB/s): The average number of MiB per second that are read from the back-end storage resources.
Back-End Write Data Rate (MiB/s): The average number of MiB per second that are written to the back-end storage resources.
Total Back-End Data Rate (MiB/s): The average rate at which data is transmitted between the back-end storage resources and the component. The rate is measured in MiB per second and includes both read and write operations.
Back-End Read Response Time (ms/op): The average number of milliseconds for the back-end storage resources to respond to a read operation.
Back-End Write Response Time (ms/op): The average number of milliseconds for the back-end storage resources to respond to a write operation.
Overall Back-End Response Time (ms/op): The average number of milliseconds for the back-end storage resources to respond to a read or a write operation.
Back-End Read Transfer Size (KiB/op): The average number of KiB that are transferred per read operation from the back-end storage resources.
Back-End Write Transfer Size (KiB/op): The average number of KiB that are transferred per write operation to the back-end storage resources.
Overall Back-End Transfer Size (KiB/op): The average transfer size, in KiB, of a read or a write operation to the back-end storage resources.

Other data

You can create performance reports that include the following information:

HPF Read I/O Rate (ops/s): The average number of read operations per second that are issued by the High Performance FICON feature of the storage system.
HPF Write I/O Rate (ops/s): The average number of write operations per second that are issued by the High Performance FICON feature of the storage system.
Total HPF I/O Rate (ops/s): The average number of I/O operations per second that are issued by the High Performance FICON feature of the storage system. This value includes both read and write operations.
HPF I/O Percentage: The percentage of all I/O operations that are issued by the High Performance FICON feature of the storage system.
PPRC Transfer Rate (ops/s): The average number of tracks per second that are transferred to the secondary device of a Peer-to-Peer Remote Copy (PPRC) pair. This value shows the activity for the source of the PPRC relationship, but shows no activity for the target.
Cache Holding Time (s): The average number of seconds that I/O data for a storage system node is held in the cache. A short cache-holding time indicates adverse performance.