IBM C2B 2.25 Brick On Sled carrier 128-pin HPC, SSA manual Idle Time Functions

Page 47

 

 

U S E R RESPONSIBLE F O R V E R I F Y I N G VERSION A N D COMPLETENESS

 

 

 

O E M F U N C T I O N A L S P E C I F I C A T I O N

U L T R A S T A R X P ( D F H C ) SSA M O D E L S

1.12/2.25 G B - 1.0" H I G H

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Notch #

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

User bytes / logical

1

2

 

3

4

5

 

6

7

8

9

10

block

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

256

20

20

 

20

19

19

 

17

17

17

17

15

 

 

 

 

 

 

 

 

 

 

 

 

 

512

13

13

 

13

12

12

 

11

11

10

10

10

 

 

 

 

 

 

 

 

 

 

 

 

 

520

12

12

 

12

12

11

 

11

10

10

10

10

 

 

 

 

 

 

 

 

 

 

 

 

 

522

12

12

 

12

12

11

 

11

10

10

10

9

 

 

 

 

 

 

 

 

 

 

 

 

 

524

12

12

 

12

12

11

 

11

10

10

10

9

 

 

 

 

 

 

 

 

 

 

 

 

 

528

12

12

 

12

12

11

 

11

10

10

10

9

 

 

 

 

 

 

 

 

 

 

 

 

 

600

11

11

 

11

11

10

 

10

10

9

9

9

 

 

 

 

 

 

 

 

 

 

 

 

 

688

10

10

 

10

10

9

 

9

9

9

8

8

 

 

 

 

 

 

 

 

 

 

 

 

 

744

9

9

 

9

9

9

 

9

8

8

8

7

 

 

 

 

 

 

 

 

 

 

 

 

 

Note: Contact an IBM Customer Representative for values at other formatted block lengths.

Table 9. Track (or Head) Skew for several block lengths

3.6 Idle Time Functions

The execution of various functions by the drive during idle times may result in delays of commands requested by initiators. ‘Idle time’ is defined as time spent by the drive not executing a command requested by a initiator. The functions performed during idle time are:

1.Servo R u n Out Measurements

2.Servo Bias Measurements

3.Predictive Failure Analysis (PFA)

4.Channel Calibration

5.Save Logs and Pointers

6.Disk Sweep

The command execution time for commands received while performing idle time activities may be increased by the amount of time it takes to complete the idle time activity. The messages and data exchanged across the SSA link are not affected by idle time activities.

Note: Command Timeout Limits do not change due to idle time functions.

All Idle Time Functions have mechanisms to lessen performance impacts for critical response time periods of operation. And in some cases virtually eliminate those impacts from an Initiator's point of view. All Idle Time Functions will only be started if the drive has not received a SCSI command for at least 5 seconds (40 seconds for Sweep). This means that multiple SCSI commands are accepted and executed without delay if the commands are received by the drive within 5 seconds after the completion of a previous SCSI command. This mechanism has the benefit of not requiring special system software (such as issuing SCSI Rezero Unit commands at known & fixed time intervals) in order to control if and when this function executes.

Note: Applications which can only accommodate Idle Time Function delays at certain times, but can not guarantee a 5 second re-instruction period, may consider synchronizing idle activities to the system needs through use of the L I T F bit in Mode Select Page 0, and the Rezero Unit command. Refer to the Ultrastar X P ( D F H C ) S S A Models Interface Specification for more details

Following are descriptions of the various types of idle functions, how often they execute and their duration. Duration is defined to be the maximum amount of time the activity can add to a command when no errors occur. N o more than one idle function will be interleaved with each command.

Following the descriptions is a summary of the possible impacts to performance.

Source filename=PERFORM

I B M Corporation

Page 47 of 87

Image 47
Contents Form Factor Disk Drive Version BM Corporation Preface Source filename=STSSHEXT When Operating Limits ReliabilityContaminants Acoustic Levels Source filename=STSSHEXT Description FeaturesGeneral Features Interface Controller FeaturesModels Reliability FeaturesGeneral SpecificationsSectors/track sct/trk User bytes/sector ub/sctSectors/logical block sct/lba User bytes/logical block ub/lbaFor Each Notch Capacity EquationsFor Entire Drive + 1 2 Volts Supply Power Requirements by Model1 C1x Models Input Voltage + 5 Volts SupplyExample 1. Calculate the mean 12 volt average current Power Calculation ExamplesExample 3. Power Calculation Example 4. Calculate the 12 volt peak current Things to check when measuring 12 V supply currentVolt current during read/write operations C1x Models Typical 12 volt current C1x Models Typical 12 volt spin-up current C1x Models 2 C2x Models Power Calculation Examples Example 4. Calculate the 12 volt peak current Volt current during read/write operations C2x Models Typical 12 volt current C2x Models Typical 12 volt spin-up current C2x Models 3 C4x Models Power Calculation Examples Example 4. Calculate the 12 volt peak current Volt current during read/write operations C4x Models Typical 12 volt current C4x Models Typical 12 volt spin-up current C4x Models 4.2 DC/DC Converter Power supply methodsCxB Models Hot plug/unplug support Grounding Requirements of the Disk EnclosureM Corporation Event Nominal Maximum Bring-up Sequence and Stop TimesBring-up Sequence Times and Stop Time for C2x Models Source filename=PERFORM # of segments Ub/lba + PerformanceEnvironment Definition Workload DefinitionBasic Component Descriptions Command Execution TimeSequential RandomData Transfer to/from Disk Data Transfer to/from SSA Link CommentsTheoretical Data Sector Transfer Rate Approximating Performance for Different EnvironmentsWhen Read-ahead is Enabled When Read Caching is EnabledWhen Write Caching is Enabled When Adaptive Caching is EnabledBack-To-Back Commands When No Seek is RequiredFor Queued Commands Reordered CommandsTrack to Track Skew SkewCylinder to Cylinder Skew Idle Time Functions Channel Calibration Servo Run Out MeasurementsServo Bias Measurements Predictive Failure AnalysisDisk Sweep Command Timeout LimitsSave Logs and Pointers M Corporation Clearances Weight and DimensionsMechanical Small Form Factor Models CxCDimensions are in millimeters Location of Side Mounting Holes of C4C Models Source filename=MECHANIC Unitized Connector Locations Source filename=MECHANIC Carrier Models CxB Dimensions CxB Models M Corporation Handle Docking and Ejection System Auto-docking Assembly Side Rails Side Rail Positioning E D Locations front view CxB Models M Corporation Electrical Connector and Indicator LocationsElectrical Interface SSA Unitized ConnectorPin Carrier ConnectorRow SSA Link Cable SSA Link Electrical CharacteristicsOption Pins and Indicators Ground long Option Port Pin Device Activity Pin/Indicator Option Port PinDevice Fault Pin/Indicator Option Port Pin Write Protect Option Port PinProgrammable pin 1 Option Port Pin Front Jumper ConnectorEarly Power Off Warning or Power Fail Power Port Pin 12 12V Charge and 5V Charge Power Port pin 1Spindle Synchronization Over Sync Hard-wire M Corporation Source filename=RELIABLE Reliability Error DetectionSeek Error Rate Power On Hours ExamplesProduct Life Power on/off cyclesUseful Life Maximum on/off cyclesSample Failure Rate Projections Install Defect FreeMean Time Between Failure *MTBF Spql Shipped product quality levelESD Protection Periodic MaintenanceConnector Insertion Cycles Install Defect Free percentage PercentOperating Limits Temperature Measurement PointsP F E T Maximum ReliabilityVibration and Shock Temperature Measurement Points for all Models bottom viewOperating Vibration Drive Mounting GuidelinesOutput Vibration Limits V4S Nonoperating VibrationNonoperating Shock ContaminantsOperating Shock Acoustic Levels Upper Limit Sound Power Requirements Bels for C4x ModelsSource filename=STANDARD Electromagnetic Compatibility EMC SafetyStandards M Corporation Serial Storage Architecture S S a P H Transport Bibliography