System (Array) Metrics
The Unisphere Performance Management application enables the user to gather, view, and analyze performance data to troubleshoot and optimize the storage systems (arrays).
Within the PM metrics section, V4 is used to denote that the metric is supported only by a storage system (array) running PowerMaxOS 10 (6079). V3 is used to denote that the metric is supported only by a storage system (array) running HYPERMAX OS 5977 or PowerMaxOS 5978. Metrics without the denotation are supported by all storage systems (arrays).
The following table lists all metrics that are available for arrays.
| Metric | Description | ||
|---|---|---|---|
| % Cache WP | The percentage of system cache that is write-pending. | ||
| % Disk Busy | The percentage of the disk that is busy. | ||
| % Effective Used Capacity | The percentage of usable capacity that would be used if all compressible data was compressed. | ||
| % Hit | The percentage of I/O operations performed by all the volumes, for which the read data was in cache and the write operation could be sent directly to cache without having to wait for data to be destaged from cache to the disks. | ||
| % Meta Data Backend Used | The used percentage of the allocation for backend configuration (TDAT devices) in the metadata page-file. The amount of page file cache that is needed is directly related to the number of TDATs in the system. A system with more drives has a larger PF cache size, leaving less cache left over for Host Cache slots. If additional TDATs are added to the system, the metadata backend used metric should be monitored to track the usage. | ||
| % Meta Data Frontend Used | The used percentage of the allocation for front-end configuration (TDEV devices) in the metadata page-file. If additional TDEVs are added to the front-end, the metadata front-end used metric should be monitored, to ensure that the 80% critical threshold is not exceeded. | ||
| % Meta Data Replication Used | The used percentage of the allocation for local replication (RDP) in the metadata page-file (V3) | ||
| % Meta Data System Used | The overall value of what percentage of the metadata page-file in the cache is in use. Systems should be monitored closely in terms of Metadata usage, Metadata Used is critical to the performance of the array. If over 90%, remediation steps should be considered. Persistently high metadata usage values that exceed the alarm threshold are also reported separately as part of the VMAX and PowerMax storage system dial-home functionality (V4). | ||
| % Provisioned | The percentage provisioned | ||
| % Provisioned Resources Used | Percent of currently provisioned capacity as part of maximum provisioning. | ||
| % Reads | The percentage of total read I/O operations performed each second by all the volumes (V3). | ||
| % Snapshot Resources Used | Capacity currently used by snapshots in percent (as part of snapshot metadata) | ||
| % Snapshot Saved | Percentage savings of the sum of all TDEV snapshot sizes (at the time of snapshot creation) and the TDEV snapshot allocated storage. | ||
| % Subscribed Capacity | The subscribed capacity as a percentage of the usable capacity (V4) | ||
| % Virtual Provisioning Saved | Percentage savings of the TDEV configured storage that is presented to the hosts and the TDEV allocated storage. | ||
| % Writes | The percentage of total write I/O operations performed by all the volumes (V4). | ||
| Allocated Capacity (GB) | The total allocated storage system capacity (in GB) | ||
| Allocated Capacity (TB) | The total allocated storage system capacity (in TB) (V4). | ||
| Avg Fall Thru Time | The average time that it takes a cache slot in LRU0 to be freed up. It is the average time from the first use of the contents to its reuse by another address (V4). | ||
| Avg Optimized Read Miss Size | The average optimized read miss size in kilobytes (V4). | ||
| BE lOs/sec | The total I/O from all BE directors to the disks per second (V3) | ||
| BE Read Reqs/sec | A data transfer of a read between the cache and the director | ||
| BE Reqs/sec | A data transfer of a read or write between the cache and the director. | ||
| BE Utilization | Percentage of time the BE is busy. | ||
| BE Write Reqs/sec | A data transfer of a write between the cache and the director | ||
| BTU | British Thermal Units (BTU) Heat from a System Level (calculated from DAE, Director-Board, MIBE (Interconnect)) (V4) | ||
| Cache Balance | Balance information for cache. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low. | ||
| Capacity Health Score | Capacity health score | ||
| Capacity of R1 Cache Slots in Use | Capacity R1 cache slots in use | ||
| Capacity of R2 Cache Slots in Use | Capacity R2 cache slots in use | ||
| Compression Dedup Savings Capacity | Total capacity saved due to compression and DeDup(both TDEVs and snapshots) (V4) | ||
| Copy Slot Count | Number of copy slots (V4) | ||
| Copy Slot Used Capacity/sec | Copy slot count used capacity | ||
| Critical Alert Count | Number of critical alerts | ||
| DA Balance | Balance information for DA. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system. the balance metrics should be relatively low. | ||
| Data Protection Health Score | Data Protection health score | ||
| Data Reduction Ratio | Data reduction ratio | ||
| Data Reduction Ratio on Enabled | Data reduction ratio on Enabled (V4) | ||
| Data Reduction Ratio on Reducible | Data Reduction Ratio on Reducible (V4) | ||
| Device WP Events/sec | The number of times each second that the write-pending limit for a specific volume was reached. | ||
| Disabled Capacity | Total capacity not enabled for data reduction (V4) | ||
| DX Balance | Balance information for DX. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low (V4). | ||
| EFD Balance | Balance information for EFD. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low. | ||
| Effective Resources Capacity | Maximum System(Array) Metadata Tbe in capacity units (V4) | ||
| Effective Used Capacity | Effective Used Capacity (V4) | ||
| FC Balance | Balance information for FC. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low. | ||
| FE Balance | Balance information for FE. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low (V3). | ||
| FE Hit Reqs/sec | The total requests from all front-end directors per second that were satisfied from cache (V3). | ||
| FE Read Hit Reqs/sec | The total read requests from all front-end directors per second that were satisfied from cache (V3). | ||
| FE Read Miss Reqs/sec | The total read requests from all front-end directors per second that were a read miss. A miss occurs when the requested data is not in cache (V3). | ||
| FE Read Reqs/sec | A read data transfer between the director and the cache. An I/O may require multiple requests depending on I/O size, alignment, or both. The requests rate should be either equal to or greater than the I/O rate. | ||
| FE Reqs/sec | A data transfer between the director and the cache. An I/O may require multiple requests depending on I/O size, alignment, or both. The requests rate should be either equal to or greater than the I/O rate (V3). | ||
| FE Utilization | Percentage of time the FE is busy (V3). | ||
| FE Write Hit Reqs/sec | The total write requests from all front-end directors per second that were satisfied from cache (V3). | ||
| FE Write Miss Reqs/sec | The total write requests from all front-end directors per second that were misses. A miss occurs when the write had to wait while data was destaged from the cache to the disks. | ||
| FE Write Reqs/sec | A write data transfer between the director and the cache. A I/O may require multiple requests depending on I/O size, alignment, or both. The request rate should be either equal to or greater than the I/O rate. | ||
| Hardware Config Health Score | Hardware configuration health score (V4) | ||
| Hardware Health Score | Hardware health score (V4) | ||
| Host lOs/sec | The number of host I/O operations performed each second by all volumes, including writes and random and sequential reads (V4). | ||
| Host Reads/sec | The number of host read operations performed each second by all volumes (V3). | ||
| Host Throughput | The number of host MBs written and read by all the volumes each second | ||
| Host Throughput Read | The number of host MBs read by all the volumes each second (V4). | ||
| Host Throughput Written | The number of host MBs written by all the volumes each second. | ||
| Host Writes/sec | The number of host write operations performed each second by all volumes. | ||
| Information Alert Count | Number of information alerts. | ||
| Local WP Count | Number of local cache slots that are write pending (V3) | ||
| Local WP Used Capacity | Capacity of local cache slots that are write pending | ||
| Max Effective Capacity | Minimum of Effective Physical Capacity and Effective Resources Capacity (V4) | ||
| Max Provisioning | Max Provisioning - System (TBp) in capacity units (V4) | ||
| Number of R1 Active Sessions | Number of R1 active sessions | ||
| Number of R1 Cache Slots in Use | Number of R1 cache slots in use (V3) | ||
| Number of R1 Msc Active Sessions | Number of R1 Msc active sessions | ||
| Number of R1 Msc Sessions | Number of R1 Msc sessions | ||
| Number of R2 Active Sessions | Number of R2 active sessions | ||
| Number of R2 Cache Slots in Use | Number of R2 cache slots in use | ||
| Number of R2 Msc Active Sessions | Number of R2 Msc active sessions | ||
| Number of R2 Msc Sessions | Number of R2 Msc sessions. | ||
| Optimized Read Misses/sec | Number of read requests each second performed directly from disks bypassing the cache. | ||
| Optimized Throughput Read Miss | Number of host MBs read each second directly from disks bypassing the cache. | ||
| Overall Efficiency Ratio | Ratio of the sum of all TDEVs + snapshot sizes (calculated based on the 128K track size) and the Physical Used Storage (calculated based on the compressed pool track size). | ||
| Overall Health Score | Overall health score | ||
| Pattern Detection Savings | Total capacity saved due to pattern detection (V4) | ||
| Performance Health Score | Performance health score | ||
| Physical Capacity | Physical capacity | ||
| Power Usage | Power consumption data, reported in Watts. (V4) | ||
| Prefetched Tracks/sec | The number of tracks per second prefetched from disk to cache upon detection of a sequential read stream. | ||
| Provisioned Capacity | Provisioned Capacity (V4) | ||
| Queue Depth Utilization | A weighted sum of the utilizations according to the number of IOs in each bucket. Queue depth range metrics are used in the calculation.Calculation: Sx=1..9 [(Avg Queue Depth Range X - 1) / Avg Queue Depth Range X] * (Queue Depth Count Range X/Queue Depth Count Total) * 100. | ||
| RDF Balance | Balance information for SRDF. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low. | ||
| RDF Utilization | Percentage of time SRDF is busy (V4). | ||
| RDFA WP Capacity | SRDF/A WP capacity | ||
| RDFA WP Count | The number of SRDF/A writes pending. (V3) | ||
| Read RT | The average time that it took the disk to serve one read command. | ||
| Reducible Capacity | Total reducible capacity (both TDEVs and snapshots) (V4) | ||
| Response Time | The overall average response time for reads and writes | ||
| SATA Balance | Balance information for SATA. Balance metrics measure hardware components balance. It is calculated as the difference between the maximum and minimum values for "% Busy" for the component instances. In a balanced system, the balance metrics should be relatively low. | ||
| SCM Balance | Difference between greatest and smallest % busy values for SCM drives (V3) | ||
| Service Level Health Score | Service level health score | ||
| Snapshot Capacity (GB) | Total snapshot capacity for system data (V3) | ||
| Snapshot Capacity (TB) | Total snapshot capacity for system data (V3) | ||
| Snapshot Compression Ratio | The ratio of the RDP Logical Backend Storage (calculated
based on the 128K track size) and the RDP Physical Used Storage of the
RDP space (calculated based on the compressed pool track size) (V3).
|
||
| Snapshot Efficiency Ratio | Ratio of the sum of all snapshot sizes and the RDP Logical Backend Storage (V3). | ||
| Snapshot Modified Capacity (GB) | Total snapshot modified capacity for system data (in GB) | ||
| Snapshot Modified Capacity (TB) | Total snapshot modified capacity for system data (in TB) | ||
| Snapshot Resources Capacity | Max Snapshot Metadata Capacity in capacity units (V4) | ||
| Snapshot Resources Used | Capacity that is used by Snapshots as part of Snapshot Metadata (V4) | ||
| Snapshot Shared Ratio | Ratio of the Snapshot Allocated Storage and the RDP Logical Backend Storage. Sum of all modified tracks on all snapshots/Sum of the RDP size for all snapshot (V4). | ||
| Software Health Score | Software health score | ||
| Subscribed Allocated Capacity (GB) | Allocated subscribed capacity (in GB) (V3) | ||
| Subscribed Allocated Capacity (TB) | Allocated subscribed capacity (in TB) (V3) | ||
| Subscribed Capacity (GB) | Subscribed capacity (in GB) | ||
| Subscribed Capacity (TB) | Subscribed capacity (in TB) | ||
| System Max WP Limit | The percent of the target % at which writes are delayed. The range is from 40% to 80%. (V3) | ||
| System Max WP Limit Capacity | System max WP limit capacity | ||
| System WP Count | The number of system cache slots that are write-pending (V3) | ||
| System WP Events/sec | The number of times each second that write activity was heavy enough to use up the system limit set for write tracks occupying the cache. When the limit is reached, writes are deferred until data in the cache is written to disk (V3). | ||
| System WP Used Capacity | WP used capacity (V3) | ||
| Total Number of R1 Sessions | Total Number of R1 sessions (V3) | ||
| Total Number of R2 Sessions | Total Number of R2 sessions (V3) | ||
| Total Number of RDF Sessions | Total Number of SRDF sessions (V3) | ||
| Unreducible Capacity | Total unreducible capacity (both TDEVs and snapshots) (V4) | ||
| Usable Capacity (GB) | The total capacity of all Storage Resource Pools (SRP) in GB (V3) | ||
| Usable Capacity (TB) | The total capacity of all Storage Resource Pools (SRP) in TB (V3) | ||
| Used Usable Capacity (GB) | Total user used capacity in GB (V3) | ||
| Used Usable Capacity (TB) | Total user used capacity in TB (V3) | ||
| Version WP Count | Version write-pending count (V3) | ||
| Version WP Used Capacity | Version write-pending used capacity (V3) | ||
| Virtual Provisioning Compression Ratio | The ratio of the TDEV Logical Backend Storage (calculated based on the 128K track size) and the TDEV Physical Used Storage (calculated based on the compressed pool track size)
|
||
| Virtual Provisioning Efficiency Ratio | The ratio of the TDEV configured storage and the TDEV Logical Backend Storage (calculated based on TDEV allocated tracks without shared unowned) | ||
| Virtual Provisioning Shared Ratio | Ratio of the TDEV allocated storage and the TDEV Logical Backend Storage. Backend Storage is the sum of TDEV allocated tracks without counting the TDEV shared unowned tracks. | ||
| Warning Alert Count | Number of warning alerts | ||
| Write RT | The average time that it took the disk to serve one write command. |
