HP StorageWorks Performance Advisor XP Software user guide Part number: T1789-96025 First edition: September 2007
Legal and notice information © Copyright 1999, 2007 Hewlett-Packard Development Company, L.P. Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor’s standard commercial license. The information contained herein is subject to change without notice.
Contents About this guide . . . . . . . . . . . . . . . . . . . . . . . . . Intended audience . . . . . . . . Prerequisites . . . . . . . . . . Related documentation . . . . . . Document conventions and symbols Rack stability . . . . . . . . . . HP technical support . . . . . . . Customer self repair . . . . . . . Product warranties . . . . . . . . Subscription service . . . . . . . HP websites . . . . . . . . . . . Documentation feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . Using the Alarm History screen . . . . . . Viewing and sorting data . . . . . . . Components of the Alarm History screen Buttons and drop-down menus . . . Columns . . . . . . . . . . . . 6 Event Log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 Database Configuration . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . Using the Database Configuration screen . . . . . . Purging performance data . . . . . . . . . . Reducing the database size . . . . . . . . . . Configuring the database size . . . . . . . . . Event logging . . . . . . . . . . . . . . . . Backing up, restoring, and resetting the database Database backup . . . . . . . . . . . . Database restore . . . . . . . . . . . . Database reset . . . . .
Logging report details and exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . Components of the View/Generate Reports screen . . . . . . . . . . . . . . . . . . . . . 16 Schedules . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . Using the Schedules screen . . . . . . Creating a schedule . . . . . . . Enabling email notifications . . . . Deleting a schedule . . . . . . . Components of the Schedule screen 17 Performance Estimator . . . . . . . . . .
D Appendix D . . . . . . . . . . . . . . . . . . . . . . . . . . 153 E Appendix E . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 I Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Supportability matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Understanding charts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chart start and end time . . . . .
Figures 1 ..HP StorageWorks CommandView XP screen . . . . . . . . . . . . . . . . . . . . 20 2 ..Grid Display query screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3 ..Grid Display screen 22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 ..RAID Group Information dialog box . . . . . . . . . . . . . . . . . . . . . . . 23 5 ..THP Pool Occupancy Information . . . . . . . . . . . . . . . . . . . . . . . . . 23 6 ..Port Data dialog box . . . . . . . . . . . . . . . .
39 ..Users tab on Security screen . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 40 ..Groups tab on Security screen . . . . . . . . . . . . . . . . . . . . . . . . . . 99 41 ..Create New User . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 ..Group Membership details for a new user . . . . . . . . . . . . . . . . . . . . . 100 100 43 ..General tab on Properties screen . . . . . . . . . . . . . . . . . . . . . . . . . 101 44 ..Members tab on Properties screen . . . . . . . .
81 ..Continuous Access Journal group utilization . . . . . . . . . . . . . . . . . . . . 150 82 ..xpdevicemap command . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 83 ..Command Device file name appended with (GUID) . . . . . . . . . . . . . . . . . 152 84 ..Charts Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 85 ..Chart plotted for all the arrays having data . . . . . . . . . . . . . . . . . . . . 156 86 ..
Tables 1 ..Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 ..Relationship between GLPR, CUs and LDEVs . . . . . . . . . . . . . . . . . . . . 14 22 3 ..Grid Display screen drop-down menus . . . . . . . . . . . . . . . . . . . . . . 24 4 ..XP48/128 (1 ACP Pair) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 5 ..XP 48/128 (2 ACP Pairs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 6 ..XP 256/512/1024 . . 7 ..XP 12000 . . . . . . 8 ..
About this guide This guide provides an overview of the main screens of Performance Advisor XP (PA XP). Each chapter describes the main functions of each screen so that you can use PA XP effectively and efficiently.
Document conventions and symbols Table 1 Document conventions Convention Element Blue text: Table 1 Cross-reference links and email addresses Blue, underlined text: http://www.hp.
WARNING! To • • • • • reduce the risk of personal injury or damage to equipment: Extend leveling jacks to the floor. Ensure that the full weight of the rack rests on the leveling jacks. Install stabilizing feet on the rack. In multiple-rack installations, fasten racks together securely. Extend only one rack component at a time. Racks can become unstable if more than one component is extended. HP technical support For worldwide technical support information, see the HP support website at: http://www.hp.
• • • • http://www.hp.com http://www.hp.com/go/storage http://www.hp.com/service_locator http://www.hp.com/support/manuals Documentation feedback HP welcomes your feedback. To make comments and suggestions about product documentation, send a message to storagedocs.feedback@hp.com. All submissions become the property of HP.
1 HP StorageWorks Performance Advisor XP 3.3 Overview HP StorageWorks Performance Advisor XP (PA XP) 3.3 is an Internet application that collects and monitors real-time performance of HP StorageWorks XP disk array products. You can use the simple, browser-based interface of Performance Advisor XP to customize performance data collection and set performance alarms.
HP StorageWorks Performance Advisor XP 3.
2 Using Performance Advisor XP Launching Performance Advisor XP HP StorageWorks Performance Advisor XP (PA XP) can be used as a standalone application or with HP StorageWorks Command View XP. To start using the application and viewing data, complete the following steps: 1. Log in to PA XP by typing the following address in a web browser: http(s)://[server name]/pa. NOTE: The PA XP URL is case-sensitive. 2. Enter the user name and password for PA XP.
Figure 1 HP StorageWorks CommandView XP screen NOTE: If you are accessing PA XP through CV XP, then you need not log into PA XP again. Refreshing Performance Advisor XP screen To refresh the information in PA XP, log out of PA XP and log in again. To obtain the latest data on the Configuration screens, click the Configuration tab on the main window of PA XP.
3 Grid Display Introduction The Grid Display screen offers a snapshot view of LDEVs and component performance data. Using this screen, you can view and monitor a variety of data about your array(s). Use the drop-down menus in the Grid Display query screen, as shown in Grid Display query screen, to refine the results displayed in the Grid Display screen. Or you can select all the required items in the fields (default). By default an Array ID is selected.
After you click Request Data, the system retrieves the requested data and displays the results of your query in the Grid Display screen. To view the details of the Grid Display query screen components, see Table 11. You can have 16384 logical devices (LDEVs) per array in a XP12000 and XP10000. To increase the number of LDEVs use, Global Logical Partition (GLPR). You can have four GLPRs per array, each GLPR has 64 control units (CUs), and each CU has 256 LDEVs, making a total of 16384 LDEVs per GLPR.
browser with information about the item. Most of the hot links when clicked display a chart of the metrics; all other hot links display information in different formats. These display formats are discussed in detail below. Red text indicates that the system has not received sufficient configuration data to display a chart. Double-click an Array ID to display the Utilization Summary screen for that array, as shown in Utilization Summary.
Figure 6 Port Data dialog box Click IO/s Chart or MB/s Chart to display the Performance History screen. You can view the data in any of the following Chart Style from the drop-down menu: • bar chart • line chart • stackable chart The Grid Display screen will not display hot links for concatenated ACP pairs in the ACP Pair ID and ACP Pair Util fields. Hence no chart for the same can be created.
Using the Summary by drop-down menu You can use the Summary by drop-down menu to view a summary of each RAID group or port on a given array. This feature totals the LDEV I/Os, LDEV MBs, cache fast write, disk fast write, cache bypass, backend transfer values, for all of the LDEVs on a given RAID group or port, and presents the data in a single line.
Figure 8 Summary by Port When the field is in Summary mode, filtering is not allowed. However, you can set filters before you select the Summary by option. For example, to view a summary of only array 10033, select that array in the Show Array drop-down menu and then select Port Summary to view summary lines for only those ports that exist on array 10033. Summary by Port shows an example of a Summary by Port screen.
Figure 9 SLPR detail view In SLPR detail view, the line above the table indicates the hierarchical information of the selected SLPR. In this figure, SLPR0 belongs to disk array serial number 10229. The SLPR view includes two tabs: CLPR and Ports. The CLPR tab displays cache size, write pending, sidefile usage, and read hits for each CLPR of the selected SLPR. Click a column heading to order the table by that column. Click an ID value to display the detail view for that CLPR.
Figure 11 CLPR detail view In CLPR detail view, the line above the table indicates the hierarchical information for the selected CLPR. In this figure, CLPR2 belongs to disk array serial number 10227 and SLPR1. The CLPR detail view displays cache size, write pending, sidefile usage, and read hits summary data for the selected CLPR. It also displays the LDEV IOs, LDEV MBs, backend transfer data, and the overall RAID group utilization percentage for each RAID group of the selected CLPR.
Continuous Access Journal Group detail view Double-click a Journal group volume ID in the Jrnl Grp column to open the Journal Group Detail View screen, as shown in Journal Group detail view and Journal Status. A list of LDEVs configured in the Journal volume displays; a maximum of 16 LDEVs display.
NOTE: The cards are lettered A-M, omitting I. Table 5 XP 48/128 (2 ACP Pairs) B, L ACP Pair 1 ACP B = 0; L = 4 F, K ACP Pair 2 ACP F = 1; K = 5 C, G CHIP Pair 1 Chip C = 0; G = 4 D, J CHIP Pair 2 Chip D = 1; J = 5 Table 6 lists the ACP and CHIP pairs for XP256/512/1024. NOTE: The cards are lettered A-Z, omitting I and O; with 12 cards on the back (A-M) and 12 on the front (N-Z).
Table 7 XP 12000 A, M ACP Pair 1 ACP A = 0; M = 4 B, N ACP Pair 2 ACP B = 1; N = 5 L, X ACP Pair 3 ACP L = 2; X = 6 K, W ACP Pair 4 ACP K = 3; W = 7 E, Q CHIP Pair 1 Chip E = 0; Q = 4 F, R CHIP Pair 2 Chip F = 1; R = 5 G, T CHIP Pair 3 Chip G = 2; T = 6 H, U CHIP Pair 4 Chip H = 3; U = 7 A, M CHIP Pair 5 Chip A = 8; M = 12 B, N CHIP Pair 6 Chip B = 9; N = 13 L, X CHIP Pair 7 Chip L = 10; X = 14 K, W CHIP Pair 8 Chip K = 11; W = 15 Table 8 lists the ACP pairs for XP10000 a
Table 9 XP24000 1AU/2MU 1AL/2ML DKA Set 1 DKA AU=0 , AL=1, DKA MU=8, ML=9 1BU/2NU 1BL/2NL DKA Set 2 DKA BU=2, BL=3 DKA NU=10,NL=11 1LU/2XU 1LL/2XL DKA Set 3 DKA LU=4, LL=5 DKA XU=12,XL=13 1KU/2WU 1KL/2WL DKA Set 4 DKA KU=6, KL=7 DKA WU=14,WL=15 1EU, 2QU 1EL, 2QL CHA Pair 1 CHA EU=0, EL=1 CHA QU=8,QL=9 1FU,2 RU 1FL,2 RL CHA Pair 2 CHA FU=2,FL=3 CHA RU=10,RL=11 1GU, 2TU 1GL, 2TL CHA Pair 3 CHA GU=4,GL=5 CHA TU=12,TL=13 1HU,2UU 1HL,2UL CHA Pair 4 CHA HU=6,HL=7 CHA UU=14,UL=15 1AU,2MU 1A
To view all the LUNs that are connected to a given CHP port, sort by CHP Port ID (and Array ID). The table groups all LUN/LDEVs that belong to the same CHP port. NOTE: If an XP1024 or XP128 array has more than 256 LUNs on a port, then PA XP may not report the additional LUNs. To sort a column in alphabetical or numerical order, click the column heading. Click the column heading again to reverse the sort order. (The columns are sorted in ascending order, by default.
Components of the Grid Display screen Table 12 list the components and menus of Grid Display screen. Double-click the blue text in the Grid Display screen, to view additional information about the fields. Table 12 Grid Display screen components Component Description ACP Pair ID The card letters for the Array Control Processor pair. ACP Pair Util The percentage of the ACP pair processors usage, during the reporting period. Array ID (Array identifier) The identification number of the array.
Component Description Device File The name of the device file. NOTE: If an array is connected to a host agent that is running on HPUX-11.31 operating system, the DSF (device special file) is displayed in a new format. A legacy DSF is displayed in parenthesis next to the new format. DFW Normal Count Disk Fast Write Normal Count. These are counts relevant to mainframe volumes. This field is zero when the emulation type starts with OPEN.
Component Description LDEV ID The identification number for the LDEV. LDEV IO/s The total I/Os per second for all random reads, random writes, sequential reads, and sequential writes during the reporting period. LDEV MB/s The total megabytes per second for the LDEV. Load Inhibit Count Count in Cache Load Inhibit Mode. Luse Indicates that the LDEV is a LUN Size Expansion (Luse).E = Master C = Component Luse Status Logical Unit Size Expansion. The CU:LDEV LUSE, when applicable.
Component Description SIO Write Cache Hits The sequential write cache hit values in per second. SS ID The identification number of the subsystem. Show ACP Pair (drop-down menu) Provides the option to display information for a particular ACP Pair or requested ACP Pairs. Show Array (drop-down menu) Provides the option to display information for the requested arrays or for one array. Show CHP Port (drop-down menu) Provides the option to display information for the requested ports or for one port.
Component Description % RGUtil Seq Read The sequential read utilization percentage for a RAID group. % RGUtil Seq Write The sequential write utilization percentage for a RAID group. % RGUtil Seq Write Parity The sequential write parity utilization percentage for a RAID group. * The following errors can occur. These errors do not prevent the collection of performance data. • erro: (short for error) This message appears when the RMLIB call that is used to display the data fails.
4 Array View Introduction Use the Array View screen to monitor the activity of specified array components in real-time. The Array View tree on the left side of the screen provides a hierarchical view of the XP arrays. Clicking on the text or adjacent icons expands the view to provide further information about the array as shown in Array View screen. Figure 15 Array View screen Information is divided into the sections: Array, CHIP Port Activity, and ACP Pair Backend.
• • • • Serial Number: The serial number of the XP array Microcode Version: The array firmware version of the XP array RmLib Version: The Raid Manager Library version installed on the host machine* DKC Time: The timestamp of when the latest performance data collection for the DKC component completed • Raid Group Time: The timestamp of when the latest performance data collection for the Raid group component/s completed • Port Time: The timestamp of when the latest performance data collection for the Port co
Array View screen. It can also have two additional optional CHIPs, B and E. These optional CHIPs can have only CHIP MPs. SVS200 is external storage and does not have any internal disks. NOTE: The mix CHIP shows only eight ports and four MPs though there are eight MPs on that board. The remaining four behave as ACP MPs. The XP24000 disk array has 32 CHIPs, 8 ACP pairs, and 4 MPs per port. The XP20000 has 8 CHIPs, 4 ACPs and 4 MPs per port.
10 busiest LDEVs/Ports To view the 10 busiest LDEVs and Ports for the selected array, click the Array Frontend Total group box on the Array View screen. The details for LDEVs and Ports are displayed in a new browser window, as shown in 10 busiest front end LDEVs and 10 busiest front end Ports. Figure 17 10 busiest front end LDEVs Figure 18 10 busiest front end Ports If the number of components are less than ten or if the utilization for them is zero, the tabs show less than 10 components.
Figure 19 10 busiest backend LDEVs Figure 20 10 busiest backend RAID groups If the number of components are less than ten or if the utilization for them is zero, the tabs show less than 10 components. The 10 busiest LDEVs selected is based on the Backend Transfer metrics and the 10 busiest RAID groups selected is based on the Overall % RAID Group Utilization metrics.
Figure 21 Cache usage chart The chart displays the cache usage, cache writes pending, and the sidefile usage measured in (MBs per 5 seconds) for the selected array. You can select the date and time range using Date/Time Filters, the maximum data points displayed from the drop-down menu and the chart style from the Chart style drop-down menu. Refreshing information Information for the Array View screen is retrieved from the database every 4 minutes and displayed in the tables.
Table 13 Array View screen components Component Description ACPs Array Control Processors. ACP Pair Backend Array Control Processor pair. Array Frontend Total The sum of all CHIP port activities. Array Backend Total The sum of all the ACP pair activities. Bus/Path Utilization percentage of the CHIP and ACP pair for the Cache Memory Bus (referred to as the Control Bus for the XP256) and Shared Memory Bus (referred to as the Data Bus for the XP256).
Array View
5 Alarm History Introduction The Alarm History screen on the Events tab displays data compiled from the database. This screen provides a history of XP array alarm events, as shown in Alarm History screen. Trapped alarms are alarms that the system has generated over time and has stored in the database. You can view trapped alarm information, including type, level, time, and status.
• • • • Email errors: The system returns anything that is non-zero in the Email Status column. SNMP errors: The system returns anything that is non-zero in the SNMP Status column. Script errors: The system returns anything that is non-zero in the Script Status column. All Errors: The system returns anything that is non-zero in the Email and Unmapped Character Select All in this column to show all items that were sent unsuccessfully.
Viewing and sorting data Each column of the table can be resized by width. Use the scrollbar at the bottom of the table if more data is available than can be seen in a single screen. Click the column headings to sort by that topic. The default sort order is ascending order. Click the column heading again to sort by descending order. Columns are sorted by alphabetical, numerical, or date order, depending on the type of information that appears in the column.
Table 14 Alarm History screen buttons and Drop-Down menus Component Description Continuous Update(button) Click to receive automatic updates of information from the database. Updates occur every five minutes. DKC/Group(drop-down) A list of the arrays that the system is monitoring or the groups that have generated alarms. Error Status(drop-down) Several selections are available to monitor error status.
Columns Table 15 Alarm History screen columns Component Description Array Type The following types are arrays are available: XP512, XP48, XP256, XP128, XP1024, XP10000, XP12000, XP24000, XP20000 and Unknown. DKC/Grp A list of the arrays, or the name of the group that generate alarms. Email Status SNMP Status These fields show whether the alarm information sent by Email or SNMP was successful, or if errors occurred. If errors occurred, a list of values is displayed for the error type(s).
Alarm History
6 Event Log Introduction The Event Log screen on the Events tab tracks errors and configuration changes, such as changing how long performance data is retained in the database, when data collection was started or stopped, or the details of an alarm configuration. The Event Log screen contains Time, Type, Severity, and Description fields. These fields that display the type of changes that are made and the time when the changes occurred.
• Description: A description of the event. Double-click the text in the Description column to open the Event Summary screen. (If the Event Summary screen does not automatically appear, it might be hidden behind the Event Log screen. Check the task bar at the bottom of your screen.) The information that appears in the Event Summary screen depends on the text that you double-clicked.
Figure 26 Event Log Calendar screen Select the start and end dates and time, and then click OK. Click Cancel to return to the main screen. Filtering by type Use the Type drop-down menu to filter the information by the type of event that occurred. Table 16 lists the types of that events available.
Table 17 Filter Severity Levels Severity Description 1, 2 Debugging messages 3, 4 Normal system operations 5, 6 User-instigated activities (for example, if the user enters the settings that instigate a purge) 7, 8 System errors 9, 10 Natural disasters For example, to receive a report on only those items that experienced system errors or worse, select 7 in this field.
Table 18 Event Log screen components Component Description Description A description of the configuration change or error. Severity (drop-down menu) The severity level of the event. 1, 2 = Debugging messages 3, 4 = Normal system operations 5, 6 = User-instigated activities (for example, if a user enters the settings that instigate a purge) 7, 8 = System errors 9, 10 = Natural disasters Refresh (button) Click to refresh the Event Log screen with current information from the database.
Event Log
7 Host Information Introduction The Host Information screen is located in the Configuration tab in PA XP as shown in Host Information screen. Use this screen to gather information from a specified host for a given array. You can use this information gathered to update the management station database.
Restoring changes If you decide not to make any modifications, click to highlight the appropriate line, and then click Restore Changes. This restores the original information. However, you cannot restore changes if you have already clicked Submit. Unrequesting information on a host To unrequest information on a host, complete the following steps: 1. Select from the list box, the host ID on which you do not want data to run.
3. Click Submit to send the request to the database. The host is removed. Viewing updated information After a request is received by the database, information is run on the selected host. After the information is run, the requested field is set to False and the received field is set to True. Click Submit to see the updated fields. Collecting data from a reconfigured array To collect data from a reconfigured array, complete the following steps: 1.
Host Information
8 Data Collection Configuration The Data Collection Configuration screen is located in the Configuration tab in Performance Advisor XP (PA XP), as shown in Data Collection Configuration screen. Use this screen to configure a command device to collect configuration and performance data for a given array. You can collect data on the performance of RAID groups, disk controller units (DKC), and ports.
NOTE: You must collect configuration data before collecting performance data for the first time. You cannot stop Configuration Data Collection once it is started. Collecting configuration data You can collect configuration data of an array either through the In Band or the Out Band mode. The in band mode uses the Raid Manager Library (RMLib) to collect data of an array. The out band mode uses the Remote Method Invocation (RMI) to collect data of the array from the SVP.
NOTE: You have a choice to restart the configuration data collection before it has completed, if you are collecting the data through the in band mode. NOTE: Performance Advisor reports only those XP Thin Provisioning and Snapshot volumes in an array that are assigned to a pool. A Snapshot volume that is not assigned to a pool but has a path assigned from a port, is grouped under VVOL-Grp. For more details, see Array View.
Figure 29 Create/View Schedule 5. To view an existing schedule, select a schedule in the Schedule drop-down box. To create a new schedule, enter a name for the schedule in the Enter Schedule Name text box. 6. Check Stagger Schedule to stagger the data collection time. If Stagger Schedule is unchecked, and Frequency is set to 15 minutes, performance data collection every quarter, irrespective of when the schedule is created. For example, if the schedule is created at 10.12 A.M.
NOTE: If you split the External RAID Group, RAID Group, XP Thin Provisioning pool, and Snapshot into two schedules, you will not see the Total I/O, Total MB/s and Total Tracks fields in the Grid Display screen. 8. Depending on what you select in the Components drop-down list, the Components list box is populated. (If you select the component RAID GROUP, a list of RAID groups configured for the array is listed.
NOTE: HP recommends that you set the data collection rate to one hour or less because of management station performance and field rollover. PA XP collects performance data on all LDEVs in the array that are reported to the management station by their respective hosts. The hosts that display their status as Received in the Array Information screen constitute the superset of mapped LDEVs. Performance data collection is not limited to the number of LDEVs that the host station is mapped to use.
If you are collecting performance data for the first time on the array, it takes longer than usual for PA XP to get the data. Subsequent performance data collection on the array is as per the time specified in the Frequency text box. NOTE: To ensure optimal performance and results, the maximum number of LDEVs for data collection is 6000. This is the total number of LDEVs across all arrays that are visible to a particular host agent.
RUNJAVA="/opt/sanmgr/jre/bin/java -DHAGT=1 -Djava.net.preferlPv4Stack=true" to RUNJAVA="/opt/sanmgr/jre/bin/java -DHAGT=1 -Djava.net.preferlPv4Stack=true -Xms100M -Xmx256M" For the host agent on Windows hosts To increase the maximum number of LDEVs for the host agent on a Windows host, complete the following steps: 1. Open the file: \Program Files\Hewlett-Packard\sanmgr\hostagent\ config\hostagentService.cfg 2. Edit the following line: JAVA_OPTIONS="Djava.security.policy=config/sec.
Table 21 Data Collection Configuration screen components Component Description Array IP Address (Text box) The IP address of the array when configuration data is collected through the out band mode. Collect Configuration Data Select to start collecting configuration data on the selected command device. Collect Performance Data Select to start collecting performance data on the selected command device.
Data Collection Configuration
9 Alarm Configuration Introduction The Alarms Configuration screen is located in the Configuration tab in Performance Advisor XP (PA XP) as shown in Alarm Configuration screen. Alarms are conditions that alert designated personnel or programs about a system malfunction, a detected condition, or an error. The Alarm Configuration screen is located under the Configuration tab. Use this screen to create and configure alarm types for specified performance metrics on an XP disk array.
• Specify a specific script to be executed when an alarm is received The Alarm Configuration screen contains several fields and buttons, as shown in Alarm Configuration screen. Figure 30 Alarm Configuration screen List box Table 22 lists the components of the list box in the Alarm Configuration screen. The screen contains the following columns that show alarm configuration templates and alarm types.
Table 22 Alarm Configuration screen columns Component Description Mod An asterisk (*) appears when an alarm is modified. N indicates new alarms. E indicates that an error has occurred. Active YES appears if the alarm is active. NO appears if the alarm is inactive. DKC The identification number of the disk controller (DKC) that is monitored. Item The LDEV or port number. Metric Category The metric category (as specified in the Chart screen) of the alarm type.
Figure 31 Add/Remove Alarm screen Use this screen to perform the following tasks: • Specify either the serial number of the array or the group name in the Array/Group drop-down box. NOTE: When you select a custom group in the Array drop-down list, will not be able to view the metrics of the External Raid Groups (Ext-RdGp), XP Thin Provisioning (THP), Snapshot, and Continuous Access Journal (Jrnl Grp) components. • Specify the category that you want to monitor by using the Category drop-down menu.
NOTE: The following metrics can be added only when an array is selected in the Array/Group drop-down box: • SM CHIP Bus/FBus Hi Util • SM ACP Bus/FBus Lo Util • CM CHIP Bus/MBus Hi Util • CM ACP Bus/MBus Lo Util • Cache Usage • Cache Write Pending Usage • Cache Sidefile Usage NOTE: If you select an item in the Metric drop-down box, pertaining to volumes, the CU drop-down menu displays. The CU drop-down menu displays LDEVs by CU.
Figure 32 Edit Schedule/Trend dialog box If you specify a level-type alarm, an email is sent only when an alarm is triggered during the specified times. For example, if you select Sunday, Mountain time, between 9 a.m. and 4 p.m., you are notified only when an alarm is triggered during this time interval. If you specify a trend-type alarm, you can see the average of change for one week, each week represented by one data point. Minimum data of two weeks is necessary to view details of a trend-type alarm.
4. Go to Start —> Programs —> HP StorageWorks —> Restart services, to restart the Performance Advisor XP service. This restarts the HpssApache, HpassPaSolid, and HpssPaTomcat services. NOTE: You must restart the service every time you modify the serverparameters.properties file.
Changing the threshold level for dispatching alarms To change the threshold level for dispatching alarms, complete the following steps: 1. Select the row(s) that you want to configure for the alarm. 2. In the Apply Action drop-down menu, click Change dispatch level on selected items. 3. In the Dispatch at Level drop-down menu, select 1, 2, or 3. 4. Click Apply Action. 5. Click Commit/Refresh to commit your changes to the database.
Figure 33 Edit Schedule/Trend window 3. Use the drop-down menus to select the start time, end time, and time zone. NOTE: Use the Duration (Week) drop-down menu only when setting trends. 4. Use the checkbox to specify the frequency of occurrence (Daily, Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, or Saturday). 5. Click OK when you are finished with your changes and proceed to step 6, or click Cancel to exit the screen without making any changes. 6.
5. Click Commit/Refresh to save your changes to the database. The location of the batch file appears in the Script column in the list box. Components of the Alarm Configuration screen Table 23 lists the components that appear on the Alarm Configuration screen. Use these components to configure alarm types. Some fields may not be available. To enable the fields or checkboxes, click the associated buttons.
Component Description Email Destination Specify the pathname of the email destination where you want to receive new alarm event information. NOTE: You must enter an email address for an alarm to be active and to receive reports. NOTE: The most common email error is because the SMTP server is not defined. For more information about this feature, see the HP StorageWorks Performance Advisor XP Software installation guide.
Alarm Configuration
10 Database Configuration Introduction The Database Configuration screen is located in the Configuration tab in Performance Advisor XP (PA XP) as shown in Database Configuration screen. Use this screen to purge performance data, specify how much disk space you want the database to use, log events at or above a specified severity level, and set the time zone of the management station.
Figure 35 Database Configuration calendar screen After you select a date, click Purge. A Confirmation dialog box appears with the following message: “Click OK to confirm deletion of records. Click Cancelto cancel.” Reducing the database size To reduce the database size after purging, complete the following steps: 1. Stop all PA XP services from: Start-> Programs-> HP Storageworks-> Stop services. 2. In the command prompt go to the directory \HPSS\padb\tools and execute the command DefragmentDB.bat. 3.
NOTE: To adjust the cap placed on the database size, see “Expand the database size to retain up to 32 GB of data” in HP StorageWorks Performance Advisor XP Software installation guide. 2. Click Save to save your change to the database. Event logging To log events that occur at or above a specified level of severity, complete the following steps: 1. Use the drop-down menu to select a number from 5 to 10, with 10 representing the highest level of severity. The default setting is 5.
Database reset At the command line prompt, enter: cd ...\Hpss\padb\tools In the Tools directory, enter: DBRestore. Components of the Database Configuration screen Table 25 lists the components that appear on the Database Configuration screen. Use these components to configure the database. Table 25 Database Configuration screen components Component Description All data for array The serial number of the array from which you want to purge data.
11 Group Configuration Introduction The Group Configuration screen is located in the Configuration tab in Performance Advisor XP (PA XP) as shown in Group Configuration screen. Use this screen to group host and array components, and to view or make comparisons between specific components. For example, if you have a database that is spread across three LDEVs, or if you want to view all LDEVs assigned to a particular port, you must use the Group Configuration screen.
1. In the Arrays list, select the array for your group. Select , or select specific arrays. To select more than one array, click each array that you want to include in your group. 2. In the Hosts list, select the host for your group. Select , or select specific hosts. To select more than one host, click each host that you want to include in your group. 3. Select the ACPs, ports, and RAID groups for your group by using the respective drop-down menus. 4.
Figure 37 Group Details window Components of the Group Configuration screen Table 26 lists the components that appear in the Group Configuration screen. Use these components to configure your groups. Table 26 Group Configuration screen components Component Description ACPs A list of available Array Control Processor Pairs. When creating your group, select all ACPs or individual ACPs in this drop-down menu. Array ID The serial number of the array. Array Type The type of array, such as XP256.
Component Description Groups A list of the groups that are created. Group Details Click to display the characteristics of a particular group. Host Group The user-defined name of the host group. Host ID The user-defined name of the customer’s machine that is connected to the array. Hosts A list of the available hosts. Jrnl Group The identification number of the Continuous Access Journal group. LDEV The identification number of the logical device. LUSE Master Logical Unit Size Expansion.
12 Data Grid Update Introduction The Data Grid Update screen is located in the Configuration tab in Performance Advisor XP (PA XP) as shown in Data Grid Update screen. Use this screen to view fabricated or incomplete records and to modify them. Fabricated records contain no host-to-array connectivity data. The Data Grid Update screen accumulates and displays LDEVs that have unknown host connections. It does not display LUSE components that are fabricated internally.
• LUN: between -1 and 15 • SSID: >= -1 If a value -1 is displayed in any of the columns, you may want to edit the column and submit it to the database. You can also use the Data Grid Update screen to modify or delete records in the database. The following section describes how to modify or delete records. Modifying records To modify a record in the Data Grid Update screen, complete the following steps: 1. Place the cursor in the field where you want to make changes and double-click.
Table 27 Data Grid Update screen components Component Description ACP Pair The identification number for the Array Control Processor Pair. Array The serial number of the array. Deleted An X appears in this column for records that have been deleted. Device File The number of the device file. Emulation An array group is divided into open volumes of the same size. These volumes are referred to as emulation types. Host The name of the host.
Data Grid Update
13 Security Introduction The Security screen is located in the Configuration tab in Performance Advisor XP (PA XP), as shown in Users tab on Security screen. It displays information about users who are authorized to use PA XP and the groups to which users are assigned. NOTE: This screen is available if PA XP has been installed as a standalone product and not if it has been installed along with CV XP. The screen consists of two tabs: Users and Groups.
Figure 39 Users tab on Security screen The Users tab displays a list of user names, which group the user is a member of, and a description of the user. By default there are two kinds of users: administrator and user. The user storageadmin is an administrator user of CV XP and has the same privileges as the administrator user in PA XP. The administrator has the highest privileges, with access to all menu options in PA XP and read and write access to all options.
Figure 40 Groups tab on Security screen Information about the members of the groups is available in the Properties window. Using the Users tab Administrators can use the Security screen to add and remove PA XP users and change passwords to enhance security for the system. General users in a user group can change only their own password and view but not edit the other information in the Security tab. Creating a User To create a new user, complete the following steps: 1. Click New in the Users tab.
Figure 41 Create New User 2. Enter the name of the new user and description for the new user in the respective fields. 3. Enter a password for the new user, and confirm it by entering it again. 4. Click Next to go to the Group Membership screen to define the membership for the user. The group membership screen appears, as shown in Group Membership details for a new user. Figure 42 Group Membership details for a new user 5. Using the arrow keys, define the membership for the user.
4. Click OK. The database is updated with the new password. Deleting a User All users except the default users, user and administrator, can be deleted. To delete a user, complete the following steps: 1. Select the user from the list box in the Users tab in the Security screen. 2. Click Delete. 3. Click Yes in the pop-up screen to delete the user. Viewing group properties To view the properties of a group, complete the following steps: 1. Click the Groups tab in the Security screen. 2. Click Properties.
Figure 44 Members tab on Properties screen 4. Click Close to return to the Security screen.
14 Charts Introduction Use the Charts screen in Performance Advisor XP (PA XP) to view the metrics of the array components in a chart format. You can compare the array components based on and within the following categories: • • • • • IO Metrics MB Metrics Utilization Metrics Backend Metrics Response Time Metrics as shown in Charts screen.
4. If you select any item in the Metric drop-down box, pertaining to volumes, the CU drop-down menu displays. The CU drop-down menu lists LDEVs by CU. Selecting a CU displays the list of LDEVs, belonging to that CU, in the Resources list box. NOTE: If you select All from the CU to plot the chart, the combined data points from both schedules is displayed. Repeated timestamps are displayed if the collection frequency for both the schedules is the same.
Figure 46 Chart Calendar screen 11. In the calendar, as shown in Chart Calendar screen, select the start and end times to filter the data. For a summary of all items in the Selected Metrics field, use the Aggregate data by drop-down menu. When you finish making your selections. Click OK to continue, or click Cancel to close the calendar without making any changes. 12.
Table 28 Chart Screen Selections for XP10000/XP12000/XP24000/XP20000 Cache Partition charts Array/Group Metric Category Metric Resource List Title Resource List Contents Any disk array model or serial number MB Metrics or Utilization Metrics Cache Usage Cache Partition All An XP 12000 serial number MB Metrics or Utilization Metrics Cache Write Pending or Cache Sidefile Usage Cache Partition All, 0, 1, 2, 3, ...
NOTE: The following metrics can be added only when an array is selected in the Array/Group drop-down box: • SM CHIP Bus/FBus Hi Util • SM ACP Bus/FBus Lo Util • CM CHIP Bus/MBus Hi Util • CM ACP Bus/MBus Lo Util • Cache Usage • Cache Write Pending Usage • Cache Sidefile Usage NOTE: When you select a custom group in the Array/Group drop-down list, will not be able to view the metrics of the External Raid Groups (Ext-RdGp), XP Thin Provisioning (THP), Snapshot, and Continuous Access Journal (Jrnl Grp) compon
Metric Category Metric Description IO Metrics, cont.
Metric Category Metric Description THP Pool Total IO THP Pool Total Random IO THP Pool Random Reads THP Pool Random Read Cache Hits THP Pool Random Writes THP Pool Total Sequential IO THP Pool Sequential Reads THP Pool Sequential Read Cache Hits THP Pool Sequential Writes THP Pool Search/ Read Basic Mode THP Pool Search/Read Hits Basic Mode THP Pool Write Basic Mode THP Pool Search/Read Seq Access Mode THP Pool Search/Read Hits Seq Access Mode THP Pool Write Seq Access Mode Each Thin Provisioning (THP)
Metric Category 110 Charts Metric Description Ext-RAID Group Total IO Ext-RAID Group Total Random IO Ext-RAID Group Random Reads Ext-RAID Group Random Read Cache Hits Ext-RAID Group Random Writes Ext-RAID Group Total Sequential IO Ext-RAID Group Sequential Reads Ext-RAID Group Sequential Read Cache Hits Ext-RAID Group Sequential Writes Ext-RAID Group Search/ Read Basic Mode Ext-RAID Group Search/Read Hits Basic Mode Ext-RAID Group Writes Basic Mode Ext-RAID Group Search/Read Seq Access Mode Ext-RAID Gr
Metric Category Metric Description MB Metrics Total MB Total Random MB Random MB Read Random MB Write Total Sequential MB Sequential MB Read Sequential MB Write Single LDEVs. Maximum Port MB Minimum Port MB Average Port MB Single port. RAID RAID RAID RAID RAID RAID RAID Group Total MB Group Total Random MB Group Random Read MB Group Random Write MB Group Total Sequential MB Group Sequential Read MB Group Sequential Write MB Each RAID group value is the sum of the LDEVs in the RAID group.
Metric Category 112 Charts Metric Description Ext-RAID Group Total MB Ext-RAID Group Total Random MB Ext-RAID Group Random Read MB Ext-RAID Group Random Write MB Ext-RAID Group Total Sequential MB Ext-RAID Group Sequential Read MB Ext-RAID Group Sequential Write MB Each External RAID group value is the sum of the LDEVs in the selected Ext-RAID groups of the array. The chart displays only one Ext-RAID Group, if configuration data is collected through the in band mode.
Metric Category Metric Description Utilization Metrics ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP ACP For example, ACP Pair BH has B as the left ACP card and H as the right ACP card.
Metric Category Metric Description Utilization Metrics, cont. Cache Usage Util Cache Writes Pending Util Cache Sidefile Usage Util Cache Usage Utilization, Cache Writes Pending Utilization, Cache Sidefile Usage Utilization THP Pool Occupancy The usage rate of a XP Thin Provisioning (THP) pool. The THP Pool Occupancy is displayed only for XP24000. SNAPSHOT Pool Occupancy The usage rate of a Snapshot pool. The Snapshot Pool Occupancy is displayed only for XP 24000.
NOTE: If a chart is not produced after you click Show Chart, the time lines of the selected metrics might not align. In other words, it is possible that old data values (for example, values from one week ago) are not aligning with new data values (for example, values from yesterday). Other possible causes are that no data was collected for the time specified, or that data was not collected from the DKC. Check the Java Console for messages. Click View, and then click Java Console in your browser.
• Start Updating: Click this button, and the system regularly checks the database for new data. If new data is found, the newest data point is added to the right side of the graph. All other data points are shifted one column to the left, and the far left column is removed. NOTE: If the chart contains metrics from different arrays, Start Updating is disabled. It is also disabled for aggregate items. Use the Chart Style drop-down menu to select the type of display in the report.
Table 30 Charts screen components Component Description Add Click to show selected metrics in a chart. Add Trend Click to show the rate of change of the selected metrics in a chart. Array/Group A list of the arrays and custom groups about which you can receive information. Clear Click to clear all selections made on the Charts screen. CU Lists the CU numbers. You must select a metric pertaining to volumes to display the CU drop-down list.
Table 31 Performance History screen components Component Description Back/Forward (buttons) Use these four buttons to display data points not currently displayed on the chart. These buttons become available when the total data points retrieved exceed the maximum data points displayed setting. Bar Chart This view shows parallel bars that are graphed side by side. The parallel display is useful for comparing items within the same time period. Select two- or three-dimensional views.
15 View/Generate Reports Introduction The View/Generate Reports screen in the Reports tab enables you to view or generate the following types of reports: array performance, LDEV IOs, cache utilization, CHIP utilization, ACP utilization, Continuous Access Journal, XP Thin Provisioning pool occupancy, and Snapshot pool occupancy. For sample reports, see Appendix B. For details on the arrays supporting XP Thin Provisioning (THP), Snapshot and Continuous Access Journal, see Appendix D.
Creating a report NOTE: You must be a member of the Administrators or StorageAdmins User group to create a report. For information about the User Groups, see the chapter Security. To create a report, complete the following steps: 1. On the View/Generate Reports screen, click New to generate a new report.
2. Select the serverparameters.properties file and open it using a text editor. 3. Search for the field SMTP_Rpt_Server and enter one or more SMTP server names. See the comments in the serverparameters.properties file for more information. 4. Go to Start —> Programs —> HP StorageWorks —> Restart services, to restart the Performance Advisor XP service. This restarts the HpssApache, HpassPaSolid, and HpssPaTomcat services. NOTE: You must restart the service every time you modify the serverparameters.
Table 32 View/Generate Reports screen components Component Description Report Name The name of the report file. Array The serial number of the disk array for which report was generated. Report Type The type of report (Array Performance, LDEV IO, Cache Utilization, CHIP Utilization, ACP Utilization, Continuous Access Journal, XP Thin Provisioning pool occupancy, and Snapshot pool occupancy. User Name The user name of the person who generated the report.
16 Schedules Introduction The Schedules screen on the Reports tab enables you to schedule report generation at periodic intervals for an array. For scheduling the reports, choose the array, the report type, and the time. The frequency can be set to daily, weekly, or monthly. The report type options are reports for array performance, LDEV IOs, cache utilization, CHIP utilization, ACP utilization, Continuous Access Journal, XP Thin Provisioning pool occupancy, and Snapshot pool occupancy.
NOTE: You must be a member of the Administrators or StorageAdmins User group to schedule reports. For more information about the User Groups, see Security. Creating a schedule Figure 52 Creating a schedule To create a schedule, complete the following steps: 1. In the Schedules screen, click New. A pop-up screen appears with the following fields, as shown in Creating a schedule: Report information • Report Type: The type of report to be generated. You can select the report from the drop-down list.
• Daily: Select if you want the report generated daily. Data of the past 24 hours is taken for daily reports. • Weekly: Select the day of the week you want the report to be generated. Data of the past 7 days is taken for the weekly reports. • Monthly: You can either select a particular day of the month or a particular week day of the month. Data of the past 30 days is considered for monthly reports. • Day of every month: Enter the day of the month. • The week and day: Select the number and day.
Table 33 Schedule screen components Component Description Array The serial number of the disk array for which report schedule is being configured. Report Type* The type of report (Array Performance, LDEV IO, Cache Utilization, CHIP Utilization, ACP Utilization, Continuous Access Journal, XP Thin Provisioning pool occupancy, and Snapshot pool occupancy. Occurrence The date or day of schedule along with the time. User Name The user name of person who configures the schedule.
17 Performance Estimator Introduction Use the Performance Estimator screen to determine the optimal performance of your arrays based on the loads that you specify by configuring the RAID groups. Subsystem configuration parameters can be rapidly entered, and the results are accurate to within 20%. Performance Estimator focuses on RAID groups, rather than on LDEVs or LUNs. NOTE: Performance Estimator does not support the XP10000/XP12000/SVS200/XP24000/XP20000 array.
Figure 54 Performance Estimator XP256/512/48 Performance Estimator features explains the Performance Estimator screen.
Figure 55 Performance Estimator features NOTE: For best results, use Internet Explorer on Windows, and use Netscape Navigator 7.0 on HP-UX. Using the Performance Estimator screen To use Performance Estimator, complete the following steps: 1. Select the desired array in the Array List drop-down menu. 2. Select items in the RAID drop-down menus. After you make your selections, a reading appears.
Components of the Performance Estimator screen Table 34 lists the components that appear in the Performance Estimator screen. Table 34 Performance Estimator screen components Component Description ACP Pairs The number of Array Control Processor (ACP) pairs. CHIP Pairs The number of Client Host Interface Processor (CHIP) pairs. Disk Type The number of gigabytes available on the drive. Low Latency Refers to a light load with less delay. Moderate Latency Refers to a heavier load with greater delay.
A Appendix A Storage management logical partitions (SLPRs) A disk array can be shared with the multiple organizations and with multiple departments within an enterprise. Therefore, multiple administrators might manage a single disk array. This circumstance creates the potential for an administrator to destroy volumes of other organizations, and it can complicate and increase the difficulty of managing the disk array.
Cache logical partitions (CLPRs) When one disk array is shared with multiple hosts, and one host reads or writes a large amount of data, the host’s read and write data occupies a large area in the disk array’s cache memory. In this situation, the I/O performance of other hosts decreases because the hosts must wait to write to cache memory. To prevent this situation, CLPR partitions the disk array’s cache memory.
B Appendix B Reports Performance Advisor supports report generation for the following categories: • • • • • • • • Array performance reports LDEV IO reports Cache utilization reports ACP utilization reports CHIP utilization reports XP Thin Provisioning (THP) pool occupancy Snapshot pool occupancy Continuous Access Journal group utilization The reports can be generated in the HTML and the PDF formats. NOTE: You must install the Acrobat Reader to view reports in the PDF format.
Figure 58 Total I/O Rate Total Back-end I/O may be compared to Total I/O and the difference is due to the effects of the array cache. The Back-end I/O load is borne by the Array Groups and ACP pairs, whereas the Total IO load is borne by the CHIP ports. NOTE: If there are no data points available for the dates selected, a blank chart will be displayed. If all the data values are zero for the dates selected, a chart with a horizontal line along X axis will be displayed in the center.
NOTE: If there are no data points available for the dates selected, blank chart will be displayed. If all the data values are zero for the dates selected, a chart with a horizontal line along X axis will be displayed in the center. Total I/O Rate Detail report The Total I/O Rate Detail report shows in a chart format, the number of Sequential I/O, Random I/O, and CFW I/O operations per second, over the entire data collection period.
Figure 61 Read/Write Ratio For example, the data point of X on the graph indicates X% read activity and (100-X)% of write activity. NOTE: If there are no data points available for the dates selected, blank chart will be displayed. If all the data values are zero for the dates selected, a chart with a horizontal line along X axis will be displayed in the center.
Figure 62 Read/Write Ratio by hour of day NOTE: If there are no data points available for the dates selected, blank chart will be displayed. If all the data values are zero for the dates selected, a chart with a horizontal line along X axis will be displayed in the center. Read/Write Detail report The Read/Write Detail report shows in a chart format, the total I/Os separated into different I/O types.
Figure 63 Read/Write Detail NOTE: If there are no data points available for the dates selected, blank chart will be displayed. If all the data values are zero for the dates selected, a chart with a horizontal line along X axis will be displayed in the center. LDEV IO reports LDEV I/O reports allow you to view the 8 busiest frontend and backend LDEVs and array groups, along with the I/Os on them.
Figure 64 Total Backend I/O Rate First Top 8 LDEVs NOTE: If I/Os have occurred for less then 8 LDEVS in an array, only those LDEVs which have I/Os associated will be displayed in the chart. Total Backend I/O Rate First Top 8 Array Groups report The Total Backend I/O Rate First Top 8 Array Groups report shows in a chart format, the real backend I/O rate for the busiest 8 Array Groups. This can be compared to the potential maximum throughput of the hardware.
Figure 65 Total Backend I/O Rate First Top 8 Array Groups NOTE: If I/Os have occurred for less then 8 LDEVS in an array, only those LDEVs which have I/Os associated will be displayed in the chart. Total Frontend I/O Rate First Top 8 LDEVs report The Total Frontend I/O Rate First Top 8 LDEVs report shows in a chart format, the number of I/O operations performed by the first busiest 8 LDEVs. Total Frontend I/O Rate First Top 8 Ldevs shows a sample Total Frontend I/O Rate First Top 8 LDEVs report for XP1024.
Figure 66 Total Frontend I/O Rate First Top 8 Ldevs NOTE: If I/Os have occurred for less then 8 LDEVS in an array, only those LDEVs which have I/Os associated will be displayed in the chart. Total Frontend I/O Rate First Top 8 Array Groups report The Total Frontend I/O Rate First Top 8 Array Groups report shows in a chart format, the number of I/O operations performed by the 8 busiest Array Groups.
Figure 67 Total Frontend I/O Rate First Top 8 Array Groups NOTE: If I/Os have occurred for less then 8 LDEVS in an array, only those LDEVs which have I/Os associated will be displayed in the chart.
Figure 68 Cache Utilization Cache Write Pending report The Cache Write Pending report shows in a chart format, the amount of data in the cache waiting to be written to a disk. This helps determine the amount of cache available. Cache Write Pending shows a sample Cache Write Pending report for XP1024. Figure 69 Cache Write Pending Percentage Read Hits report The Percentage Read Hits report shows in a chart format, cache read hits as a percentage of the total cache read operations.
Figure 70 Percentage read hits Total Backend Transfer report The Total Backend Transfer report shows in a chart format, the total number of transfers, sequential, random drive-to-cache, and cache-to-drive, per second. Total Backend Transfer report shows a sample Total Backend Transfer report for XP1024.
Figure 72 Total Backend Transfer by Hour of the Day Cache Side File Utilization report The Cache Side File Utilization report shows in a chart format, the cache side file utilization. The cache side file utilization is used for the software Asynchronous Continuous Access XP. It holds the data buffers that have not been acknowledged by the remote host. Cache Side File Utilization shows a sample Cache Side File Utilization report for XP1024.
ACP Utilization report The ACP Utilization report shows in a chart format, the average utilization of the ACP pairs over the entire data collection period. ACP utilization over the entire data collection period shows a sample ACP Utilization report for XP1024. Figure 74 ACP utilization over the entire data collection period ACP Utilization by Hour of the Day report The ACP Utilization by Hour of the Day report shows in a chart format, the average utilization of the ACP pairs over a 24-hour period.
Figure 75 ACP utilization over a 24-hour period CHIP utilization reports The CHIP utilization reports allow you to view in a chart format, the utilization data for all CHIP pairs in the array and the average utilization data for all the CHIP pairs in the array. A sample of each report is given below: CHIP Utilization report The CHIP Utilization report shows in a chart format, the utilization data for all CHIP pairs in the array. CHIP Utilization shows a sample CHIP Utilization report for XP12000.
Figure 76 CHIP Utilization CHIP Utilization by Hour of the Day report The CHIP Utilization by Hour of the Day report shows in a chart format, the utilization data for all CHIP pairs in the array averaged over a 24-hour period. CHIP Utilization by Hour of the Day shows a sample CHIP Utilization by Hour of the Day report for XP12000.
Figure 78 CHIP Processor Utilization In this sample report, the percentage utilization of individual processors (MP) on the CHIP CHA 1E is shown. Similarly a report will be generated for all CHIPs in the array. XP Thin Provisioning (THP) pool occupancy The THP pool occupancy report show the usage percentage of the eight busiest THP pools. NOTE: PA XP reports only those THP volumes in an array that are assigned to a pool.
NOTE: PA XP reports only those Snapshot volumes in an array that are assigned to a pool. Utilization Metric displays Snapshot Pool Occupancy only for the XP24000 and XP20000. Figure 80 Snapshot pool occupancy Continuous Access Journal group utilization The Journal group utilization report shows the utilization percentage of the eight busiest Journal groups.
C Appendix C Performance Advisor Configuration and Performance Data Collection Prerequisites for Windows Operating System To overcome device renaming and device remapping issues during data collection, it is recommended to use the Windows globally unique identifier (\\?\Volume{GUID}) to perform configuration and performance data collection instead of the using the device file names.
4. Restart PA XP hostagent services. For example, :\Program Files\Hewlett-Packard\sanmgr\hostagent\sbin> hastart.bat 5. On the management station launch Performance Advisor XP and click on the Configuration tab. Select Host Information tab and click on Request Info followed by Submit. 6. Once the Status column in the Host Information screen is updated with the status Received, click on the Data Collection Configuration tab.
D Appendix D Supportability matrix The following matrix displays the supportability of XP Thin Provisioning, Snapshot and Continuous Access Journal on the XP arrays.
Appendix D
E Appendix E Understanding charts Charts display a color-coded graph of historical data. The Charts Graph screen appears when you click on Show Charts in the Charts screen or when you double-click an LDEV IO/s, LDEV MB/s, chip Util., ACP Pair ID, ACP Pair Util., Backend Transfer, Avg Read Resp(msec), Max Read Resp (msec)-valid for last 30 secs, Avg Write Resp(msec), and Max Write Resp (msec)-valid for last 30 secs, (shown in blue text) in the Grid Display screen.
For charts with a sync error, that is when more than one graph is plotted, the start and end time is the data range of the master graph which resulted in the sync error. For Daily and Weekly charts the start and end times are the start and end day. When charts are and are not plotted Charts may or may not be plotted depending of certain factors. This section explains when charts are plotted and with what data points, and when they are not plotted.
When plotting a chart, if the reference item, or the item with the latest starting point, has only one data point, while the rest of the items have more data points, the chart will be plotted only with a single data point of the reference item, as shown in Figure 87.
Figure 89 Chart is not plotted if all the items do not have data points Exception • Chart is plotted even secondary item has no data points When there are data points for the reference item for a given time, but not for the secondary items, a chart is still plotted.
Index Symbols accessing PA XP, 19 ACP Pair summary screen, 40 Alarm Configuration screen, 73 Alarm History screen, 47 alarm notifications configuring, 80 scheduling, 80 alarms changing threshold levels, 80 configuring, 73, 79 configuring threshold levels, 79 creating, 78 disabling, 81 displaying historical data, 47 enabling, 79 scripts, 81 trapped, 47 types, 77 Appendix, 131, 133 Appendix C, 151 Appendix D, 153 Array View screen, 39 arrays charting components, 106 configuring data collection, 63 creating g
database backing up, 87 changing records, 94 configuring size, 86 logging events, 87 purging, 85 resetting, 88 restoring, 87 updating, 59 Database Configuration screen, 85 deleting, 123 groups, 90 disabling alarms, 81 disk space requirements, 70 displaying alarm historical data, 47 array activity, 39 authorized groups, 97 authorized users, 97 charts, 103 charts with different data collection rates, 114 fabricated records, 93 group properties, 90, 101 host and array components, 89 incomplete records, 93 LDEV
P Performance Advisor Configuration and Performance Data Collection Prerequisites for Windows Operating System, Performance data Collecting, 65 creating, 65 viewing, 65 performance data displaying, 21 displaying charts, 103 purging, 85 Performance Estimator screen, 127 Performance History screen, 115 performance, optimizing, 127 prerequisites, 13 Properties screen, 99 properties, displaying group, 101 purging performance data, 85 R rack stability warning, 14 RAID groups displaying details, 23 optimizing
Search HP StorageWorks Performance Advisor XP Software user guide 163
Search