Table 89 Disk Drive Lite Subprofile Key CIM Classes (continued)
Class
TPD_StoragePool
TPD_VolumeBasedOnDiskExtent
TPD_StoragePoolComponent
TPD_AssociatedPrimordialDisks
Description
Identifies which Storage Pool contains the storage provided by specific disk drives.
Associations between a StorageVolume and the
DiskStorageExtent that makes up the volume. Includes support for TPVV.
Associations between a storage pool and the
DiskStorageExtent that make up the pool. Support for DynamicStoragePool (CPG) and
DeltaReplicaStoragePool (snap space) is included.
Associations between the primordial StoragePool and the DiskStorageExtents that make up the pool.
Supported Methods
Table 90 (page 98)
lists the supported method of the Disk Drive Lite subprofile:
Table 90 Methods for TPD_StorageSystem
Method modifyInstance()
Description
Provides the ability to set owner and contact information for the StorageSystem.
Block Server Performance Subprofile
The Block Server Performance Subprofile defines classes and methods for managing performance information (statistics) in block servers.
For detailed information regarding the Block Server Performance subprofile, refer to SMI-S at http://www.snia.org
.
Block Server Performance Subprofile CIM Classes
Table 91 Block Server Performance Subprofile Key CIM Classes
Class
TPD_StorageSystem
TPD_StatisticsCollection
TPD_DiskStatisticalData
TPD_PortStatisticalData
TPD_VolumeStatisticalData
TPD_LUNStatisticalData
TPD_ArrayStatisticalData
TPD_NodeStatisticalData
Description
The storage array on which the statistics are to be collected.
A collection of statistics data.
An instance of statistics data for a TPD_DiskDrive.
An instance of statistics data for a TPD_FCPort.
An instance of statistics data for a TPD_StorageVolume.
An instance of statistics data for an exported
TPD_StorageVolume .
An instance of statistics data for the entire array.
An instance of statistics data for a particular node controller.
98 CIM API SMI-S Support
TPD_ArrayStatisticalData
Statistical data for the top level ComputerSystem. Data are aggregate of all the port statistics in the array.
Property
StatisticTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
ReadIOTimeCounter
WriteIOTimeCounter
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Type
DateTime
Unit16
Unit64
Unit64
Description
Time the statistics is collected.
Computer System (2) Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of all reads.
The cumulative count of all writes.
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from
TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
TPD_NodeStatisticalData
Statistical data for a node controller, including statistics of:
• all the ports on the node
• cached memory pages
• all CPU’s on the node
Property
StatisticTime
StartStatisticsTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Type Description
DateTime Time the statistics is collected
DateTime The time, relative to the node controller where the statistic was collected, when the first measurement was taken. If the statistic is reset, the
StartStatisticTime is the time when the reset was performed.
Unit16 Peer Computer System (4). Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of all reads.
The cumulative count of all writes.
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
Block Server Performance Subprofile 99
Property
ReadIOTimeCounter
WriteIOTimeCounter
ReadHitIOs
WriteHitIOs
ReadMissIOs
WriteMissIOs
LockedReadBlocks
LockedWriteBlocks
FreeCachePages
CleanCachePages
Write1CachePages
WriteNCachePages
WriteScheduledCachePages
WritingCachePages
PendingDelayedCopyOn
-WriteCachePages
Unit32
Unit32
Unit32
Unit32
Unit32 ProcessingDelayedCopyOn
-WriteCachePages
FC10kRPMCfcDirty Unit32
FC15kRPMCfcDirty
NLCfcDirty
SSDCfcDirty
FC10kRPMCfcMax
FC15kRPMCfcMax
NLCfcMax
SSDCfcMax
FC10kRPMDelayedAck
FC15kRPMDelayedAck
Type
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit32
Unit32
Unit32
Description
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
The cumulative count of all read cache hits (Reads from Cache).
The cumulative count of Write Cache Hits (Writes that went directly to Cache).
The cumulative count of all Read Cache misses.
The cumulative count of all Write Cache misses.
Number of pages being modified by host read I/O that are being written to disk by the flusher.
Number of pages being modified by host write I/O that are being written to disk by the flusher.
Number of cache pages without valid data on them.
Number of clean cache pages (valid data on page). A page is clean when data in cache matches data on disk.
Number of dirty pages that have been modified exactly 1 time. A page is dirty when it has been modified in cache but not written to disk.
Number of dirty pages that have been modified more than 1 time.
Number of pages scheduled to be written to disk.
Number of pages being currently written by the flusher to disk.
Number of pages waiting for delayed copy on write resolution.
Unit32
Unit32
Unit32
Unit32
Unit32
Unit32
Unit32
Unit32
Unit32
Number of pages currently being processed for delayed copy on write resolution.
Current number of dirty cluster memory pages in the system for the 10k RPM
FC disk type.
Current number of dirty cluster memory pages in the system for the 15k RPM
FC disk type.
Current number of dirty cluster memory pages in the system for the NL disk type.
Current number of dirty cluster memory pages in the system for the SSD disk type.
Maximum allowed number of dirty cluster memory pages in the system for the 10k RPM FC disk type.
Maximum allowed number of dirty cluster memory pages in the system for the 15k RPM FC disk type.
Maximum allowed number of dirty cluster memory pages in the system for the NL disk type.
Maximum allowed number of dirty cluster memory pages in the system for the SSD disk type.
Number of delayed acknowledgements to the host in order to throttle the host's IO writes due to cache resource constraints for the 10k RPM FC disk type.
Number of delayed acknowledgements to the host in order to throttle the host's IO writes due to cache resource constraints for the 15k RPM FC disk type.
100 CIM API SMI-S Support
Property
NLDelayedAck
SSDDelayedAck
TotalCPUUserTimeCounter
TotalCPUIdleTimeCounter
TotalCPUSystemTimeCounter
Interrupts
ContextSwitches
Type
Unit32
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Description
Number of delayed acknowledgements to the host in order to throttle the host's IO writes due to cache resource constraints for the NL disk type.
Number of delayed acknowledgements to the host in order to throttle the host's IO writes due to cache resource constraints for the SSD disk type.
The user time since system start time, using ClockTickInterval units, for all the
CPUs on the node.
The idle time since system start time, using ClockTickInterval units, for all the
CPUs on the node.
The system time since system start time, using ClockTickInterval units, for all the CPUs on the node.
The number of interrupts since system start time of the CPUs on the node.
The number of context switches since system start time of the CPUs on the node.
TPD_VolumeStatisticalData
Property
StatisticTime
StartStatisticsTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
ReadIOTimeCounter
WriteIOTimeCounter
ReadHitIOs
WriteHitIOs
ReadMissIOs
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
WriteMissIOs Unit64
ReadServiceTimeCounter Unit64
WriteServiceTimeCounter
Type
DateTime
DateTime
Unit16
Unit64
Description
Time the statistics is collected
The time, relative to the volume where the statistic was collected, when the first measurement was taken. If the statistic is reset, the StartStatisticTime is the time when the reset was performed.
Volume (8) Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of all reads.
The cumulative count of all writes.
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from
TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
The cumulative count of all read cache hits (Reads from Cache).
The cumulative count of Write Cache Hits (Writes that went directly to Cache).
The cumulative count of all Read Cache misses.
The cumulative count of all Write Cache misses.
The service time since system start time, using ClockTickInterval units, for the Read
I/Os.
The service time since system start time, using ClockTickInterval units, for the Write
I/Os.
Block Server Performance Subprofile 101
Property Type
TotalServiceTimeCounter Unit64
QueueLength
PercentBusy
TPD_PortStatisticalData
Unit32
Unit32
Description
The service time since system start time, using ClockTickInterval units, for the total
I/Os.
The queue length
Percent of current busy use.
Property
StatisticTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
ReadIOTimeCounter
WriteIOTimeCounter
Type
DateTime
Unit16
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Description
Time the statistics is collected.
Front-End Ports (6) or Back-End Ports (7) Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of a reads.
The cumulative count of all writes.
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from
TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
TPD_DiskStatisticalData
Property
StatisticTime
StartStatisticsTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Type
DateTime
DateTime
Unit16
Description
Time the statistics is collected.
The time, relative to the disk where the statistic was collected, when the first measurement was taken. If the statistic is reset, the StartStatisticTime is the time when the reset was performed.
Disk Drive (10) Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of all reads.
The cumulative count of all writes.
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from
TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
102 CIM API SMI-S Support
Property
ReadIOTimeCounter
Type
Unit64
WriteIOTimeCounter Unit64
ReadServiceTimeCounter Unit64
WriteServiceTimeCounter Unit64
TotalServiceTimeCounter Unit64
Description
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
The service time since system start time, using ClockTickInterval units, for the Read
I/Os.
The service time since system start time, using ClockTickInterval units, for the Write
I/Os.
The service time since system start time, using ClockTickInterval units, for the total
I/Os.
The queue length
Percent of current busy use.
QueueLength
PercentBusy
TPD_LUNStatisticalData
Unit32
Unit32
Property
StatisticTime
ElementType
TotalIOs
ReadIOs
WriteIOs
KBytesTransferred
KBytesWritten
KBytesRead
IOTimeCounter
ReadIOTimeCounter
WriteIOTimeCounter
Type
DateTime
Unit16
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Unit64
Description
Time the statistics is collected.
Front-End Ports (6) or Back-End Ports (7) Defines the role that the element played for which this statistics record was collected.
The cumulative count of I/Os.
The cumulative count of all reads.
The cumulative count of all writes.
The cumulative count of data transferred in Kbytes (1024bytes = 1KByte).
The cumulative count of data written in Kbytes (1024bytes = 1KByte).
The cumulative count of data read in Kbytes (1024bytes = 1KByte).
The cumulative elapsed I/O time (number of Clock Tick Intervals) for all I/Os as defined in 'Total I/Os'. This value can be divided by number of IOs to obtain an average response time. ClockTickInterval value can be obtained from
TPD_BlockStatisticsCapabilities. It is set to 1 microsecond.
The cumulative elapsed time for all Read I/Os for all cumulative Reads.
The cumulative elapsed time for all Write I/Os for all cumulative Writes.
Supported Methods
Table 92 Methods for TPD_BlockStatisticsService
Method
GetStatisticsCollection ()
Description
Retrieves statistics in CSV bulk format. The collection of statistics returned is determined by the list of element types passed in to the method and the manifests for those types contained in the supplied BlockStatisticsManifestCollection. If both the Elements and
BlockStatisticsManifestCollection parameters are supplied, then the types of elements returned is an intersection of the element types listed in the Elements parameter and the types for which BlockStatisticsManifest instances exist in the supplied
BlockStatisticsManifestCollection. The statistics are returned through a well-defined array of strings, whose format is specified by the StatisticsFormat parameter, that can be parsed to retrieve the desired statistics as well as limited information about the elements that those metrics describe. Please note that only CSV StatisticsFormat is supported.
Block Server Performance Subprofile 103
