White PapersThe AiO600 is a simple, integrated, and affordable network storage solution for small and mid-size companies. The AiO600 provides shared storage expansion for application servers, file serving, and comprehensive data protection, allowing companies to easily manage, grow, and protect their data.
Testing was performed with the HP AiO600 876GB serial attached SCSI (SAS) model with six (6) 146GB front load, hot plug, 15,000 RPM SAS disks attached to a Smart Array E200i controller with 128MB battery-backed write cache (BBWC) offering RAID 0, 1, 10, and 5. The AiO600 also has an embedded NC373i Multifunction Gigabit Network Adapter and an additional dual port NC7170 10/100/1000T network interface card (NIC).
ESRP storage solution brief for the HP StorageWorks AiO600 supporting 1,200 Microsoft Exchange users Overview............................................................................................................................................ 3 Disclaimer........................................................................................................................................... 3 Features.............................................................................................................................................. 3 Solution description.............................................................................................................................. 3 HP All-in-One Storage System Management ........................................................................................ 5 HP All-in-One Storage Manager (ASM)............................................................................................... 9 Microsoft iSCSI Software Target....................................................................................................... 10 HP Array Configuration Utility...................................................................................................... 12 Test environment................................................................................................................................ 12 Targeted customer profile................................................................................................................ 12 Tested deployment ......................................................................................................................... 13 Simulated Exchange configuration................................................................................................ 13 Hardware.................................................................................................................................. 14 Software ................................................................................................................................... 14 Disk configuration: Mailbox store disks ......................................................................................... 15 Disk configuration: Transaction log disks....................................................................................... 15 Disk configuration: Streaming backup to disk................................................................................. 15 Best practices.................................................................................................................................... 16 Shared storage.............................................................................................................................. 16 Array and controller configuration ................................................................................................... 16 Array expansion ........................................................................................................................ 16 Database volume configuration.................................................................................................... 16 iSCSI array configurations........................................................................................................... 17 iSCSI initiator installation ................................................................................................................ 17 Initiator configuration.................................................................................................................. 17 Target Portals............................................................................................................................. 17 Load Balanced Targets................................................................................................................ 17 Backup strategy ............................................................................................................................. 18 Test results summary........................................................................................................................... 19 Reliability...................................................................................................................................... 19 Performance.................................................................................................................................. 19 Untitled DocumentStreaming backup performance ....................................................................................................... 20Database read-only performance ................................................................................................. 20 Log read-only performance.......................................................................................................... 20 Backup to disk performance......................................................................................................... 20 Conclusion........................................................................................................................................ 21 Appendix A test reports ................................................................................................................... 22 Database checksum reliability testing............................................................................................. 22 Performance results 2 hour performance test................................................................................... 25 Backup to disk database read only................................................................................................ 27 Log replay performance soft recovery testing .................................................................................. 29 Backup to disk performance results................................................................................................ 31 For more information.......................................................................................................................... 34 Untitled DocumentOverview This document provides information on the HP StorageWorks 600 All-in-One Storage System (AiO600) storage solution for Microsoft Exchange Server 2003 (Exchange), based on the Microsoft Exchange Solution Reviewed Program (ESRP) Storage program. The ESRP Storage program was developed by Microsoft Corporation to provide a common storage testing framework for vendors to provide information on their storage solutions for Microsoft Exchange Server software. For more details on the Microsoft ESRP Storage program, please visit http://www.microsoft.com/technet/prodtechnol/exchange/2003/esrp.mspx. Disclaimer This document has been produced independently of Microsoft Corporation. Microsoft Corporation expressly disclaims responsibility for, and makes no warranty, express or implied, with respect to the accuracy of the contents of this document. Features This solution brief describes a tested configuration for deploying Exchange Server 2003 Enterprise Edition (Exchange) in a 1,200 user environment. The performance results and best practices outlined in this document provide tested guidelines for configuring Exchange using the HP StorageWorks 600 All-in-One Storage System (AiO600). The AiO600 is a simple, integrated, and affordable network storage solution for small and mid-size companies. The AiO600 provides shared storage expansion for application servers, file serving, and comprehensive data protection, allowing companies to easily manage, grow, and protect their data. Testing was performed with the HP AiO600 876GB serial attached SCSI (SAS) model with six (6) 146GB front load, hot plug, 15,000 RPM SAS disks attached to a Smart Array E200i controller with 128MB battery-backed write cache (BBWC) offering RAID 0, 1, 10, and 5. The AiO600 also has an embedded NC373i Multifunction Gigabit Network Adapter and an additional dual port NC7170 10/100/1000T network interface card (NIC). All-in-One Storage Systems provide shared storage expansion for application servers, and are designed to be deployed on the existing Ethernet network in less than 30 minutes by IT generalists with little or no storage expertise. They are targeted at the small-to-medium (SMB) businesses that need to lower the cost and complexity associated with growing storage environments. In addition to the best practices presented in this solution brief, more information on Exchange Server 2003 and HP products are available for free download from HP ActiveAnswers at www.hp.com/solutions/exchange. Solution description Information on the All-in-One Storage Systems can be found in the Products Designed for Microsoft Windows Windows Catalog and HCL here: http://www.windowsservercatalog.com/item.aspx?idItem=21a9b263-b811-6ffb-f573-3c994b680053. When deploying Exchange, it is important to not only size for storage capacity; you must also consider the storage performance demands as well. Validating that the storage subsystem is capable of supporting the anticipated workload will minimize the risk of the disk subsystem becoming an I/O bottleneck. 3 Untitled DocumentThis solution brief provides tested best practices to help administrators design and implement the AiO600 for a 1,200-user Exchange deployment. One HP ProLiant DL320 G3 server was used to drive the ESRP I/O load and simulate the production Exchange 2003 server. An additional HP StorageWorks All-in-One Storage System, the AiO400 was used to provide backup to disk for Exchange database storage. The AiO400 was configured with 4 x 250GB front load, hot plug, 7200 RPM SATA disks connected via a 4-port SATA RAID controller offering RAID 0, 1, 10, and 5. Figure 1 below shows the HP StorageWorks 600 All-in-One Storage System (AiO600) rack and tower models. Note the six (6) front load, hot plug internal SAS or SATA drive bays. The AiO600 tested was the 876GB SAS model. It is also available with an optional Tower to Rack Conversion Kit. Figure 1. HP StorageWorks 600 All-in-One Storage System (AiO600) rack and tower models The HP StorageWorks All-in-One Storage Systems come with Windows Storage Server 2003 Release 2 (R2) Standard Edition pre-installed, and also include the integrated All-in-One Storage System Management. 4 Untitled DocumentHP All-in-One Storage System Management The HP All-in-One Storage System Management reduces the time, training, expense and expertise required to set up and manage storage for applications such as Microsoft Exchange Server, SQL Server and file storage and sharing. Designed for IT generalists, not storage experts, this tool handles underlying storage tasks and presents capacity in application-centric terms. Data migration tools for Exchange and SQL Server include best practices recommendations which allow an IT generalist to operate as a storage expert. Administrators can quickly and easily expand data areas, migrate data from capacity constrained servers, implement disk and tape backup policies, setup end-user file shares, manage disk quotas and more. The Application View in HP All-in-One Storage System Management is shown below in Figure 2. This is for a newly configured system, and shows the current usage of applications such as Exchange Server, SQL Server, file storage and sharing, and also user-defined applications. Figure 2. Application View in HP All-in-One Storage System Management 5 Untitled DocumentFigure 3 below shows the HP All-in-One Storage System Management, including the options for Storage Hosting, System Settings and Status, and Utilities such as the Rapid Startup Wizard. Notice the link to Microsoft iSCSI Target under Storage Management, which will be detailed later in this document. Figure 3. HP All-in-One Storage System Management, including Storage Hosting, System Settings, Status, and Utilities 6 Untitled DocumentFigure 4 below shows the Exchange Storage Group Wizard, which allows you to remotely connect to an Exchange server from the All-in-One Storage System and relocate the Exchange files to the All-in-One Storage System. Figure 4. Using the Exchange Storage Group Wizard to relocate Exchange files to AiO600 7 Untitled DocumentFigure 5 below shows a Status warning in the HP All-in-One Storage System Management, alerting the administrator of low disk space. Figure 5. Status warning in HP All-in-One Storage System Management 8 Untitled DocumentHP All-in-One Storage Manager (ASM) Figure 3 and Figure 5 above both show the link to the All-in-One Storage Manager (ASM), which is the primary interface for configuring new storage on the AiO Storage System and migrating existing applications. ASM provides an application-centric approach that hides the complexities associated with storage provisioning and is easier to learn because it uses terms and operations familiar to application administrators. The All-in-One Storage Manager (ASM) should be consistently used as the primary interface for configuring storage on the AiO Storage System; because if other interfaces are used, storage provisioned outside of ASM cannot be managed by the HP All-in-One Storage System Management. Figure 6 below shows configuring new storage and migrating Exchange databases with the HP All-in-One Storage Manager (ASM). Figure 6. Configuring new storage and migrating Exchange databases with the HP All-in-One Storage Manager (ASM) 9 Untitled DocumentMicrosoft iSCSI Software Target The All-in-One Storage Systems include the Microsoft iSCSI Software Target to provide shared storage for an iSCSI storage area network (SAN). The Microsoft iSCSI Software Target is an industry standard iSCSI solution, which means it provides the performance of iSCSI and the low cost of using existing TCP/IP network hardware and expertise. Using a separate network subnet or isolated hardware for networked storage traffic is recommended for best performance and data security. The Microsoft iSCSI Software Target is illustrated in Figure 7 below, showing the ease of extending an iSCSI virtual disk to provide more storage to an Exchange database. Figure 7. Extending an iSCSI virtual disk in the Microsoft iSCSI Software Target 10 Untitled DocumentThe following section covers the utilities and processes used for manually configuring storage (once the application requirements are known, such as RAID type and disk size). However, the HP All-in-One Storage Manager (ASM) is the primary interface for configuring storage on the AiO Storage System and ASM reduces these complexities. If ASM is not used as the primary interface for configuring storage on the AiO Storage System, the overall manual process of creating storage for applications or file shares consists of: 1. Creating the Arrays (physical disks) and Logical Drives (RAID and Size) in the Array Configuration Utility, as illustrated below in Figure 9. 2. Creating volumes by partitioning and formatting the logical drives on the AiO600 Windows Storage System using the diskpart command line utility and the logical disk manager console, 3. Creating a Virtual Disk on the volume in the Microsoft iSCSI Software Target, and assigning this disk to an iSCSI Target (shown in Figure 7) 4. Verifying Initiator (host) access, as shown below in Figure 8, and logging into the iSCSI disk target from that host using the Microsoft Initiator. A later section covers manual configuration for load balancing targets across network adapters. Figure 8. Verifying Initiator (host) access in the Microsoft iSCSI Software Target 11 Untitled DocumentHP Array Configuration Utility Figure 9 below shows the details of the SAS Array and Logical Drives in the HP Array Configuration Utility on the AiO600. Figure 9. SAS Array and Logical Drives in HP Array Configuration Utility For more information on setting up the AiO600 for other applications and environments, see the documentation accompanying the product and the HP website links at the end of this document. Test environment The testing discussed in this report covers three major aspects critical to Exchange stability and performance including the following: " 24 Hour Stress Test measures the longer term stability of the Exchange infrastructure running under a load for a full 24-hour day s duration. " 2 Hour Performance Test provides performance under a heavy load for a shorter time frame. " Backup performance backup-to-disk testing measures the time required to perform streaming Exchange backups. Each of these three test points is critical in validating a storage subsystem for a high performing, stable Exchange environment. Targeted customer profile This solution is targeted toward a 1,200 user Exchange environment, with the following conditions: " Up to 1,200 users per AiO600 storage system " Multiple hosts can be attached to the iSCSI storage 12 Untitled Document" 0.5 I/O operations per second (IOPS) per user " 200MB mailbox quota " 2 storage groups per server " 5 databases per storage group " 120GB per storage group " Disk to Disk (D2D) backups to the AiO400 SATA disk array " Database restore service level agreement (SLA) based on 24GB database size Note The ESRP-Storage program focuses on storage solution testing to address performance and reliability issues with storage design. However, storage products are not the only factor to take into consideration when designing an Exchange solution. Other factors that affect the server performance are: server processor utilization, server physical and virtual memory limitations, resource requirements for other applications, directory and network service latencies, network infrastructure limitations and recovery requirements, and client usage profiles. All these factors are beyond the scope for ESRP-Storage. Therefore, the number of mailboxes hosted per server as part of the tested configuration may not necessarily be viable for some customer deployments. For more information on identifying and addressing performance bottlenecks in an Exchange system, please refer to Microsoft's Troubleshooting Microsoft Exchange Server Performance, available at http://go.microsoft.com/fwlink/?LinkId=23454. Tested deployment The following tables summarize the testing environment: Simulated Exchange configuration Number of Exchange mailboxes simulated 1,200 Number of hosts 1 Number of mailboxes/host 1,200 Number of storage groups/host 2 Number of mailbox stores/storage group 5 Number of mailboxes/mailbox store 120 Number of mailbox store LUNs/storage group 1 Simulated profile: I/Os per second per mailbox (IOPS, include 20% headroom) 0.5 IOPS/user Database LUN size 185 GB Log LUN size 35 GB Backup LUN size/storage group 354 GB = 177 GB per storage group Total database size for performance testing 234 GB % storage capacity used by Exchange databases 234 / 370 = 63% 13 Untitled DocumentHardware Storage type (SAN, DAS, iSCSI, NAS) iSCSI Storage model and OS/firmware revision HP AiO600 Windows Storage Server 2003 R2 2x2.67 GHz Dual-Core Intel Xeon CPU 1GB RAM Storage cache 128MB (Battery-backed) Number of storage controllers 1 (E200i integrated RAID controller 1.20A firmware) Number of storage ports 2 (Gigabit Ethernet) Maximum bandwidth of storage connectivity to host 2x 1 Gigabit Switch type/model/firmware revision HP ProCurve 2824 FW I.08.58. NIC model and firmware 2x HP NC7782 Gigabit server adapter (host) HP NC373i multifunction gigabit adapter (Storage Server) Jumbo frames Not tested1Number of NICs/host 2 Host server type HP ProLiant DL320 G3, 2.93GHz Intel Celeron D CPU 2 GB RAM Total number of disks tested in solution Database: 6 x146GB 10K SAS Logs: 2x 250GB 7.2K RPM SATA Backups: 4x 250GB 7.2K RPM SATA Maximum number of spindles can be hosted in the storage Internally, as configured: 6 x 146GB 10K SAS Can be expanded via external disk enclosures 1Note on Jumbo Frames: Not all customer environments necessarily support larger frame sizes than the default Ethernet frame size (also known as Jumbo Frames) so it was not used in this test environment. Software NIC driver 9.52.0.0 Microsoft iSCSI Software Initiator v2.02 iSCSI offload Not used Multi-Pathing Not applicable Host OS Microsoft Windows Server 2003 R2 ESE.dll file version 6.5.7638.2 Replication solution name/version None 14 Untitled DocumentDisk configuration: Mailbox store disks Disk type, speed and firmware revision SAS 10K disks HPD4 HP DF146A8B57 Raw capacity per disk (GB) 146GB Number of physical disks in test 6 Total raw storage capacity (GB) 876 GB Disk slice size (GB) Not applicable Number of slices per LUN or number of disks per LUN Not applicable RAID level RAID 1+0 striped disk mirroring Total formatted capacity (GB) 326 GB Storage capacity utilization 37% Database capacity utilization 27% Disk configuration: Transaction log disks Disk type, speed and firmware revision SATA 7.2K RPM S0 Raw capacity per disk (GB) 250GB Number of spindles in test 2 Total raw storage capacity (GB) 250 GB Disk slice size (GB) Not applicable Number of slices per LUN or number of disks per LUN Not applicable RAID level RAID 1+0 disk mirroring Total formatted capacity (GB) 35 GB x 2 Note: See the section iSCSI array configurations for discussion of sizing and placing the transaction log volume on the iSCSI array or local to the Exchange server (direct-attached storage). Disk configuration: Streaming backup to disk Disk type, speed and firmware revision SATA 7.2K RPM S0 Raw capacity per disk (GB) 250GB Number of spindles in test 4 Total raw storage capacity (GB) 1,000 GB Disk slice size (GB) Not applicable Number of slices per LUN or number of disks per LUN Not applicable RAID level RAID 1+0 disk mirroring Total formatted capacity (GB) 353 GB 15 Untitled DocumentBest practices Exchange Server 2003 is very I/O intensive and places high demands on the disk resources, thus it is important to ensure that an adequate number of fast spindles are provided for disk performance. Note With any Exchange 2003 deployment, the best practice is to test the environment with tools such as Jetstress and LoadSim to ensure that the storage subsystem is meeting the necessary performance requirements of the design. Shared storage While the storage solution presented here is sized to allow additional I/O overhead, the impact of sharing storage between Exchange and other applications needs to be fully tested by the customer before deploying. The impact of other applications will be to reduce the available performance and the number of users may need to be adjusted accordingly. It is strongly advised that the production system database disks be monitored (using Windows perfmon) for these counters: Avg. Disk sec/Read and Avg. Disk sec/Write, and that these latencies should average under 20 milliseconds. Array and controller configuration In general, isolate Exchange database and log workloads into separate logical drives on separate arrays. This will improve performance and reduce the potential risk of loss of data in the event of a disk failure. Use RAID 10 (also known as RAID 1+0) disk mirroring for increased performance and availability of the database and log LUNs. While RAID 5 may provide additional storage capacity, it also incurs additional write overhead, lowering the overall disk I/O performance of write-demanding applications such as Exchange. For the AiO600, to improve disk I/O performance, configure the E200i array controller cache for 75% write (though 50/50 read/write is the default). This was found to be the most optimal for caching of I/O in the iSCSI environment. Array expansion Like other HP Smart Array controllers, the AiO600 supports online array expansion, meaning that the Exchange database array can be sized to support a smaller number of users initially and later grown to meet higher I/O capacity, without necessarily incurring downtime. Depending on the configuration, additional external enclosures and drives could be added as needed. Best practices dictate that a backup should be made of all data and the expansion should occur outside of operating hours and sustained production load. Database volume configuration It should be noted that the Exchange Storage Group Migration wizard creates an individual logical drive for each Exchange database. As part of the testing for this ESRP, an individual logical drive was created for each Exchange Storage Group, as the performance will not be impacted by the choice of configuration. 16 Untitled DocumentiSCSI array configurations In the testing here, the transaction logs were placed on the Exchange server internal drives. Another configuration for Exchange is to place the transaction logs on the iSCSI array instead of the server internal drives; however the total load will need to be reduced or disk latencies will increase. In the testing here, the transaction logs were placed on the AiO600 and the storage solution was able to sustain over 600 IOPS but database disk latencies approach the 20ms threshold. By placing the transaction logs on the Exchange server, performance did improve and additional disk space was freed up on the AiO600. iSCSI initiator installation This environment uses the Microsoft iSCSI Software Initiator v2.02 (which is a free download from Microsoft.com). The Microsoft iSCSI Initiator software is included as part of the AiO application agent server install. If the end user installs the iSCSI software separately, the All-in-One Storage Manager (ASM) will not interoperate with it correctly. It may be necessary to uninstall and reinstall correctly using the AiO application agent server install. See the AiO startup documentation for more information. Initiator configuration Typically, the Microsoft iSCSI initiator configuration must be started before you can present Logical Drives from the array to the iSCSI host, also known as an Initiator Node. The iSCSI array needs to be made aware of the initiators to which it will be presenting targets. The first step in Initiator Configuration is to log into the iSCSI array via a target portal (combination if IP address and port number). This can be verified on the Discovery tab of the Microsoft iSCSI initiator. Target Portals The IP address of a Gigabit port on the AiO600 is added to the Initiator. These IP addresses are assigned during AiO600 configuration. Load Balanced Targets For best performance on a server with dual NICs, it is recommended that you manually assign each disk Target to a different source as described here. On the Targets tab of the Initiator, click the Refresh button and you should see your disk targets. Select a Target and click the Log on& button. The Automatically restore& option should be used for production iSCSI disks which will then be mounted when the server reboots. Click the Advanced button and configure similar to the manual configuration shown in Figure 10 assigning one NIC to one Target Portal, rather than using the Default selection for all options. For example, matching odd and even IP addresses is easiest to keep track of. In the streaming backup to disk, storage group drives could be paired with one of the two backup drives so that backup could be load balanced across both NICs. Note As a best practice for fault tolerance, create separate physical arrays for Exchange databases and logs. For best performance and load balancing of iSCSI traffic, it is important to create individual Logical Drives and assign them to different Targets in the AiO600 configuration shown in Figure 9. Each of these targets can then be assigned to a different NIC on the Exchange server. 17 Untitled DocumentFigure 10 shows the load balanced manual configuration. This is also where you would select the iSCSI Host Bus Adapter if using an iSCSI offload card such as the HP NC370T Multifunction Network Adapter or Alacritech or QLogic cards instead of the default NIC. Figure 10. Manually assigning first NIC to Target Portals for load balancing Backup strategy A well designed and implemented backup and recovery strategy should be a top priority for Exchange implementations. There are various backup and restore strategies that can be implemented, depending on the requirements of the environment. The backup-to-disk tested here is not a complete data protection strategy; as it places the backup copy on another array on the same network as the primary databases. Some method such as copying the backup data files to tape or other media at a remote site should be considered to remove the single point of failure and potential loss of data. In this solution, three tests were performed: database read-only performance, log read-only performance and backup-to-disk performance. The database read-only performance test was used to measure the read I/O performance metrics by running a checksum on all the databases and log files. This test can help determine what kind of database read throughput can be achieved during backups. The log replay test was used to measure the maximum rate at which the log files can be played against the databases. This is used to determine what type of database write throughput can be achieved during a log recovery. And the backup-to-disk performance measures the overall throughput of a disk-to-disk (D2D) type of backup using the AiO600 for databases and the AiO400 for backup storage. 18 Untitled DocumentTest results summary This section provides a high-level summary of the test data from ESRP and the links to the detailed html reports which are generated by the ESRP testing framework. Please click on the links below to view the html report for each test. ReliabilityOne of the tests in the framework is designed to test reliability over a 24-hour test period. The goal is to verify that the storage can handle high I/O load for a long period of time. Both log and database files will be analyzed for integrity after the stress test to ensure no database/log corruption. The following list provides an overview: " Any errors reported in the saved event log file? There were no errors reported in the event log for the storage reliability testing. " Any errors reported during the database and log checksum process? There were no errors reported for the database and log checksum process. " If backup to disk test is done, any errors reported during the process? There were no errors reported for the backup to disk testing PerformancePerformance testing is performed to exercise the storage with maximum sustainable Exchange type of I/O for two hours. The test is to show how long it takes for the storage to respond to an I/O under load. The data below is a sample taken from the attached host and is the average of all the logical disks in the two-hour test duration. Average of the database disks read latency (ms) 16 ms Average of the database disks write latency (ms) 14 ms Average of the log disks write latency (ms) Average database disk read IO per second 507 IOPS Average database disk write IO per second 243 IOPS Total database disk IOPS 750 IOPS Average log disk write IO per sec 59 IOPS Max database page fault stalls per sec 0 19 Untitled DocumentStreaming backup performanceFor the ESRP Version 1.2 release, only streaming backup type is supported for testing in the framework. There are three tests in this section. The first two tests measure the database and log read I/O performance metrics by running a checksum on all the databases and log files. The third test is to measure the end-to-end performance when the databases are backed up to disks. Database read-only performanceThe test is to measure the maximum rate at which databases could be streamed to backup devices. The following table shows the average values per host. MB read/sec per Storage group 54.6 MB/sec MB Read/sec total 109.2 MB/sec File size/Seconds taken 240401 MB/2201 seconds Log read-only performanceThe test is to measure the maximum rate at which the log files can be played against the databases. The following table shows the average rate for 100 log files played in a single storage group. Each log file is 5MB in size. Average time to play one Log file (sec) 4.9 sec Average log disks Read Bytes/sec 2,132,267 (2,082MB/sec) Backup to disk performanceThis test runs a streaming backup for all the database files, and stores them on disks. The following table shows an average rate at which each storage group can be backed up. Total database size per storage group (GB) 117.4 GB Time taken to backup each storage group 66 minutes Average MB backed up/sec per storage group 7.6 MB/sec Average MB backup up/sec per server 15.2 MB/sec 20 Untitled DocumentConclusion The information discussed in this solution brief highlights the tested performance results and configuration best practices for an Exchange server supporting 1,200 Exchange users. Testing was performed with the HP AiO600 876GB serial attached SCSI (SAS) model with six (6) 146GB front load, hot plug, 15,000 RPM SAS disks attached to a Smart Array E200i controller with 128MB battery-backed write cache (BBWC) offering RAID 0, 1, 10, and 5. Testing demonstrated that the storage subsystem was capable of supporting both the IOPS needed to support the 1,200 Exchange users at 0.5 IOPS each, as well as providing additional headroom. All-in-One Storage Systems provide shared storage expansion for application servers, and are designed to be deployed on the existing Ethernet network in less than 30 minutes by IT generalists with little or no storage expertise. They are targeted at the small-to-medium (SMB) businesses that need to lower the cost and complexity associated with growing storage environments. This document is developed by storage solution providers and reviewed by the Microsoft Exchange Product team. The test results/data presented in this document are based on the tests introduced in the ESRP test framework. A customer should not quote the data directly for his/her pre-deployment verification. It is still necessary to go through the exercises to validate the storage design for a specific customer environment. The ESRP program is not designed to be a benchmarking program; tests are not designed to achieve the maximum throughput for a given solution. Rather, the program is focused on producing recommendations from vendors for Exchange application. Consequently, the data presented in this document should not be used for direct or indirect comparisons among the solutions. 21 Untitled DocumentAppendix A test reports Database checksum reliability testing The following table shows a quick overview of checksum statistics: Database Pages seen Bad checksumsCorrectable checksumsWrong page numbers MB read/sec (File size/Seconds taken) E:Jetstress.edb 5128450000 54.8 MB/sec (20033 MB/365 seconds) E:Jetstress1.edb 5128450000 54.8 MB/sec (20033 MB/365 seconds) E:Jetstress2.edb 5128450000 54.7 MB/sec (20033 MB/366 seconds) E:Jetstress3.edb 5129218000 54.7 MB/sec (20036 MB/366 seconds) E:Jetstress4.edb 5128706000 54.6 MB/sec (20034 MB/366 seconds) E:Jetstress5.edb 5128706000 54.7 MB/sec (20034 MB/366 seconds) F:Jetstress.edb 5128194000 54.3 MB/sec (20032 MB/369 seconds) F:Jetstress1.edb 5128962000 54.3 MB/sec (20035 MB/368 seconds) F:Jetstress2.edb 5127426000 54.3 MB/sec (20029 MB/368 seconds) F:Jetstress3.edb 5129218000 54.4 MB/sec (20036 MB/368 seconds) F:Jetstress4.edb 5128450000 54.4 MB/sec (20033 MB/368 seconds) F:Jetstress5.edb 5128450000 55.4 MB/sec (20033 MB/361 seconds) (sum) 61542680000 109.2 MB/sec (240401 MB/2201 seconds) 22 Untitled DocumentThe following table shows a quick overview of database and log performance counter sample data: Storage Volume Avg. Disk sec/Read Avg. Disk sec/Write Disk Reads/sec Disk Writes/sec Total Seconds E: 0.5860.000874.75 0.0002197.56F: 0.5890.002871.22 0.0062205.70The following table shows a quick overview of processor and memory performance counter sample data: Counter Average Minimum Maximum % Processor Time 54.520.000 79.69Available MBytes 1621.181589.00 1654.00Free System Page Table Entries192861.50192856.00 192856.00Pages/Sec 0.0000.000 1.000Pool Nonpaged Bytes 20029610.00 20004860.00 20111360.00Pool Paged Bytes 20956930.00 20504580.00 20967420.00Performance log C:\Jetstress\DatabaseChecksum_2006_9_7_0-51-14.blg is saved. 9/6/2006 10:57:30 PM -- Environment validation results: Detected JetstressUI version: 6.5.7830.0 Detected operating system: Microsoft Windows Server 2003 R2 (5.2.3790.0) Service Pack 1 Detected ESE.dll version: 6.5.7638.2 Detected ESEPERF.dll version: 6.5.7638.1 9/6/2006 10:57:30 PM -- Validating input parameters, it may take a few minutes... 9/6/2006 10:57:30 PM -- Attaching databases ... (it may take a few minutes if the databases are in dirty shutdown state) 9/6/2006 10:57:35 PM -- Loading performance counters... 9/6/2006 10:57:35 PM -- Instance5660.1: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Instance5660.2: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Start Jetstress test... 9/6/2006 10:57:36 PM -- Starting StreamingBackup test run... 9/6/2006 10:57:38 PM -- Performance logging started. 9/6/2006 10:57:38 PM -- Performance data will be saved to C:\Jetstress\Performance_LogGeneration_2006_9_6_22-57-35.blg. 9/7/2006 12:33:38 AM -- Adding new data to the performance log file... 9/7/2006 12:33:53 AM -- Performance logging stopped. 9/7/2006 12:33:53 AM -- Stopping Jetstress... 9/7/2006 12:34:52 AM -- Creating test report ... 9/7/2006 12:34:53 AM -- Volume F: has 0.0169 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume E: has 0.0167 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume L: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume L: has 0.0019 for Avg. Disk sec/Read. 23 Untitled Document9/7/2006 12:34:53 AM -- Volume M: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume M: has 0.0013 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Test has 0 Max Database Page Fault Stalls/sec. 9/7/2006 12:34:53 AM -- Test has 0 Database Page Fault Stalls/sec samples higher than 0. 9/7/2006 12:34:53 AM -- Soft recovery may take a while depending on the number of log files. 9/7/2006 12:34:55 AM -- Performance logging started. 9/7/2006 12:34:55 AM -- Performance data will be saved to C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.blg. 9/7/2006 12:51:11 AM -- Adding new data to the performance log file... 9/7/2006 12:51:14 AM -- Performance logging stopped. 9/7/2006 12:51:14 AM -- Soft recovery is completed, please open C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.html for soft recovery result. 9/7/2006 12:51:16 AM -- Performance logging started. 9/7/2006 12:51:16 AM -- Performance data will be saved to C:\Jetstress\DatabaseChecksum_2006_9_7_0-51-14.blg. 9/7/2006 12:51:16 AM -- Checksum validation may take a while depending on the file sizes. 9/7/2006 1:28:03 AM -- Database checksum in progress: Storage Group #1 (100%), and Storage Group #2 (100%). 9/7/2006 1:28:03 AM -- Performance logging stopped. 9/7/2006 1:28:03 AM -- Checksum is completed, please open C:\Jetstress\DatabaseChecksum_2006_9_7_0-51-14.html for checksum result. 24 Untitled DocumentPerformance results 2 hour performance test The purpose of this test is to verify the storage configuration which the customer is planning to deploy. The test run has been successful, meaning the database read latency and log write latency are below 20ms; Database page fault stalls/sec is 0. However, Microsoft strongly recommends that you perform further validation of this storage solution (additional tasks listed below): 1.You should refer to the table below to determine if the actual disk IO (achieved IO) has exceeded the targeted IO (expected IO). If not, you may want to increase the thread count to increase the load, provided that the disk latency will not exceed the threshold. 2.You need to confirm whether the storage solution has synchronous replication as part of the implementation. If the solution does not utilize synchronous replication, then you should check for the database write latency, the desired value should be under 20 ms. (Please refer to http://support.microsoft.com/?kbid=895847 for synchronous replication definitions.) 3.Please check the status pane in the Jetstress window, to make sure that no errors were logged during the database checksum validation. Planned disk subsystem profile: Total test database size Production data size Total number of databases Expected I/O Achieved I/O 233.84 GB (based on the attached database) (n/a) 12 (2 storage(s) * 6 database(s)) 600.00 (1200 mailboxes of 0.50 IOPS) 750.06 JetStress test parameter summary: InstancethreadCountlogBufferSizeopInsertopReplaceopDeletelazyCommit1 490001770 5902 490001770 590Disk subsystem performance summary: Volume Avg. Disk sec/Read Avg. Disk sec/Write Disk Reads/sec Disk Writes/sec Avg. Disk Bytes/Write Data (E:) 0.0160.011255.14122.53 (n/a) Data (F:) 0.0160.016251.63120.77 (n/a) Log (L:) 0.0000.0000.00029.99 6682.39Log (M:) 0.0000.0000.00029.49 6697.6625 Untitled DocumentProcessor/memory performance summary: Counter Average Minimum Maximum % Processor Time 9.2716.667 22.37Available MBytes 726.24696.00 1568.00Free System Page Table Entries192872.00192872.00 192872.00Pages/Sec 0.2210.000 73.60Pool Nonpaged Bytes 18683805.66 18665472.00 19042304.00Pool Paged Bytes 20185720.78 19845120.00 20238336.00Database Page Fault Stalls/sec0.0000.000 0.000Performance log C:\Jetstress\Performance_2006_9_6_13-23-35.blg is saved. 9/6/2006 1:23:25 PM -- Environment validation results: Detected JetstressUI version: 6.5.7830.0 Detected operating system: Microsoft Windows Server 2003 R2 (5.2.3790.0) Service Pack 1 Detected ESE.dll version: 6.5.7638.2 Detected ESEPERF.dll version: 6.5.7638.1 9/6/2006 1:23:25 PM -- Validating input parameters, it may take a few minutes... 9/6/2006 1:23:25 PM -- Attaching databases ... (it may take a few minutes if the databases are in dirty shutdown state) 9/6/2006 1:23:36 PM -- Loading performance counters... 9/6/2006 1:23:36 PM -- Instance4024.1: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 1:23:36 PM -- Instance4024.2: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 1:23:36 PM -- Start Jetstress test... 9/6/2006 1:23:37 PM -- Starting Performance test run... 9/6/2006 1:23:40 PM -- Performance logging started. 9/6/2006 1:23:40 PM -- Performance data will be saved to C:\Jetstress\Performance_2006_9_6_13-23-35.blg. 9/6/2006 3:23:26 PM -- Adding new data to the performance log file... 9/6/2006 3:23:40 PM -- Performance logging stopped. 9/6/2006 3:23:40 PM -- Stopping Jetstress... 9/6/2006 3:24:16 PM -- Creating test report ... 9/6/2006 3:24:18 PM -- Volume F: has 0.0162 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Volume E: has 0.0159 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Volume L: has 0.0001 for Avg. Disk sec/Write. 9/6/2006 3:24:18 PM -- Volume L: has 0.0000 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Volume M: has 0.0002 for Avg. Disk sec/Write. 9/6/2006 3:24:18 PM -- Volume M: has 0.0000 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Test has 0 Max Database Page Fault Stalls/sec. 9/6/2006 3:24:18 PM -- Test has 0 Database Page Fault Stalls/sec samples higher than 0. 26 Untitled DocumentBackup to disk database read only The following table shows a quick overview of checksum statistics: Database Pages seen Bad checksumsCorrectable checksumsWrong page numbers MB read/sec (File size/Seconds taken) E:Jetstress.edb 5119746000 54.7 MB/sec (19999 MB/365 seconds) E:Jetstress1.edb 5119490000 54.8 MB/sec (19998 MB/365 seconds) E:Jetstress2.edb 5119746000 54.8 MB/sec (19999 MB/365 seconds) E:Jetstress3.edb 5120258000 54.7 MB/sec (20001 MB/365 seconds) E:Jetstress4.edb 5120258000 54.6 MB/sec (20001 MB/366 seconds) E:Jetstress5.edb 5120258000 54.7 MB/sec (20001 MB/365 seconds) F:Jetstress.edb 5120258000 54.4 MB/sec (20001 MB/367 seconds) F:Jetstress1.edb 5120002000 54.5 MB/sec (20000 MB/366 seconds) F:Jetstress2.edb 5119234000 54.5 MB/sec (19997 MB/366 seconds) F:Jetstress3.edb 5120514000 54.4 MB/sec (20002 MB/367 seconds) F:Jetstress4.edb 5119746000 54.4 MB/sec (19999 MB/367 seconds) F:Jetstress5.edb 5119746000 55.2 MB/sec (19999 MB/362 seconds) (sum) 61439256000 109.3 MB/sec (239997 MB/2196 seconds) 27 Untitled DocumentThe following table shows a quick overview of database and log performance counter sample data: Storage Volume Avg. Disk sec/Read Avg. Disk sec/Write Disk Reads/sec Disk Writes/sec Total Seconds E: 0.5860.003874.68 0.0082194.50F: 0.5870.003872.26 0.0072199.80The following table shows a quick overview of processor and memory performance counter sample data: Counter Average Minimum Maximum % Processor Time 55.080.000 82.81Available MBytes 1625.611554.00 1681.00Free System Page Table Entries192866.50192872.00 192872.00Pages/Sec 0.1290.000 99.95Pool Nonpaged Bytes 18758140.00 18669570.00 18866180.00Pool Paged Bytes 20289530.00 19722240.00 20484100.00Performance log C:\Jetstress\DatabaseChecksum_2006_9_6_15-24-18.blg is saved. 9/6/2006 1:23:25 PM -- Environment validation results: Detected JetstressUI version: 6.5.7830.0 Detected operating system: Microsoft Windows Server 2003 R2 (5.2.3790.0) Service Pack 1 Detected ESE.dll version: 6.5.7638.2 Detected ESEPERF.dll version: 6.5.7638.1 9/6/2006 1:23:25 PM -- Validating input parameters, it may take a few minutes... 9/6/2006 1:23:25 PM -- Attaching databases ... (it may take a few minutes if the databases are in dirty shutdown state) 9/6/2006 1:23:36 PM -- Loading performance counters... 9/6/2006 1:23:36 PM -- Instance4024.1: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 1:23:36 PM -- Instance4024.2: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 1:23:36 PM -- Start Jetstress test... 9/6/2006 1:23:37 PM -- Starting Performance test run... 9/6/2006 1:23:40 PM -- Performance logging started. 9/6/2006 1:23:40 PM -- Performance data will be saved to C:\Jetstress\Performance_2006_9_6_13-23-35.blg. 9/6/2006 3:23:26 PM -- Adding new data to the performance log file... 9/6/2006 3:23:40 PM -- Performance logging stopped. 9/6/2006 3:23:40 PM -- Stopping Jetstress... 9/6/2006 3:24:16 PM -- Creating test report ... 9/6/2006 3:24:18 PM -- Volume F: has 0.0162 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Volume E: has 0.0159 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Volume L: has 0.0001 for Avg. Disk sec/Write. 9/6/2006 3:24:18 PM -- Volume L: has 0.0000 for Avg. Disk sec/Read. 28 Untitled Document9/6/2006 3:24:18 PM -- Volume M: has 0.0002 for Avg. Disk sec/Write. 9/6/2006 3:24:18 PM -- Volume M: has 0.0000 for Avg. Disk sec/Read. 9/6/2006 3:24:18 PM -- Test has 0 Max Database Page Fault Stalls/sec. 9/6/2006 3:24:18 PM -- Test has 0 Database Page Fault Stalls/sec samples higher than 0. 9/6/2006 3:24:19 PM -- Performance logging started. 9/6/2006 3:24:19 PM -- Performance data will be saved to C:\Jetstress\DatabaseChecksum_2006_9_6_15-24-18.blg. 9/6/2006 3:24:19 PM -- Checksum validation may take a while depending on the file sizes. 9/6/2006 4:01:00 PM -- Database checksum in progress: Storage Group #1 (100%), and Storage Group #2 (100%). 9/6/2006 4:01:00 PM -- Performance logging stopped. 9/6/2006 4:01:00 PM -- Checksum is completed, please open C:\Jetstress\DatabaseChecksum_2006_9_6_15-24-18.html for checksum result. Log replay performance soft recovery testing The following table shows a quick overview of log replay statistics for 100 log files per storage group: Storage Volume Avg. Disk sec/Read Avg. Disk sec/WriteDisk Reads/secDisk Writes/sec Disk Read Bytes/sec Total SecondsLog (L:) 0.001 0.00032.330.376 2104680.00497.77Log (M:) 0.001 0.00033.180.458 2159853.00481.48The following table shows a quick overview of processor and memory performance counter sample data: Counter Average Minimum Maximum % Processor Time 14.450.000 88.46Available MBytes 942.83767.00 1651.00Free System Page Table Entries192856.00192856.00 192856.00Pages/Sec 0.0010.000 0.500Pool Nonpaged Bytes 19697230.00 18751490.00 20041730.00Pool Paged Bytes 21002420.00 20480000.00 21008380.00Performance log C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.blg is saved. 9/6/2006 10:57:30 PM -- Environment validation results: Detected JetstressUI version: 6.5.7830.0 Detected operating system: Microsoft Windows Server 2003 R2 (5.2.3790.0) Service Pack 1 Detected ESE.dll version: 6.5.7638.2 Detected ESEPERF.dll version: 6.5.7638.1 9/6/2006 10:57:30 PM -- Validating input parameters, it may take a few minutes... 29 Untitled Document9/6/2006 10:57:30 PM -- Attaching databases ... (it may take a few minutes if the databases are in dirty shutdown state) 9/6/2006 10:57:35 PM -- Loading performance counters... 9/6/2006 10:57:35 PM -- Instance5660.1: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Instance5660.2: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Start Jetstress test... 9/6/2006 10:57:36 PM -- Starting StreamingBackup test run... 9/6/2006 10:57:38 PM -- Performance logging started. 9/6/2006 10:57:38 PM -- Performance data will be saved to C:\Jetstress\Performance_LogGeneration_2006_9_6_22-57-35.blg. 9/7/2006 12:33:38 AM -- Adding new data to the performance log file... 9/7/2006 12:33:53 AM -- Performance logging stopped. 9/7/2006 12:33:53 AM -- Stopping Jetstress... 9/7/2006 12:34:52 AM -- Creating test report ... 9/7/2006 12:34:53 AM -- Volume F: has 0.0169 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume E: has 0.0167 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume L: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume L: has 0.0019 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume M: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume M: has 0.0013 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Test has 0 Max Database Page Fault Stalls/sec. 9/7/2006 12:34:53 AM -- Test has 0 Database Page Fault Stalls/sec samples higher than 0. 9/7/2006 12:34:53 AM -- Soft recovery may take a while depending on the number of log files. 9/7/2006 12:34:55 AM -- Performance logging started. 9/7/2006 12:34:55 AM -- Performance data will be saved to C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.blg. 9/7/2006 12:51:11 AM -- Adding new data to the performance log file... 9/7/2006 12:51:14 AM -- Performance logging stopped. 9/7/2006 12:51:14 AM -- Soft recovery is completed, please open C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.html for soft recovery result. 30 Untitled DocumentBackup to disk performance results This test is to verify the storage configuration for a streaming backup operation. This table reports the performance metrics of streaming backup for each storage group backed up. Storage Group Storage Group Size (MB) Backup Time (hh:mm:ss.msec) Average MB Backed Per Second 1 120203.05 02:11:41.75 15.21 2 120198.05 02:11:52.64 15.19 This table reports the performance metrics of streaming backup for each database backed up. Storage GroupDatabase NameDatabase Size (MB)E:Jetstress.edb 20033.01 E:Jetstress1.edb20033.01 E:Jetstress2.edb20033.01 E:Jetstress3.edb20036.01 E:Jetstress4.edb20034.01 1 E:Jetstress5.edb20034.01 F:Jetstress.edb 20032.01 F:Jetstress1.edb20035.01 F:Jetstress2.edb20029.01 F:Jetstress3.edb20036.01 F:Jetstress4.edb20033.01 2 F:Jetstress5.edb20033.01 Planned disk subsystem profile: Total test database size Production data size Total number of databases Expected I/O Achieved I/O 234.36 GB (based on the attached database) (n/a) 12 (2 storage(s) * 6 database(s)) 600.00 (1200 mailboxes of 0.50 IOPS) 485.23 Jetstress test parameter summary: InstancethreadCountlogBufferSizeopInsertopReplaceopDeletelazyCommit1 490001770 5902 490001770 59031 Untitled DocumentDisk subsystem performance summary: Volume Avg. Disk sec/Read Avg. Disk sec/Write Disk Reads/sec Disk Writes/sec Avg. Disk Bytes/Write Data (F:) 0.0170.000242.600.034 (n/a) Data (E:) 0.0160.000242.560.034 (n/a) Log (L:) 0.0000.0000.0100.108 29.17Log (M:) 0.0000.0000.0100.093 247.79Processor/memory performance summary: Counter Average Minimum Maximum % Processor Time 25.527.388 34.86Available MBytes 737.53708.00 1674.00Free System Page Table Entries192856.00192856.00 192856.00Pages/Sec 0.0050.000 0.200Pool Nonpaged Bytes 20045979.15 19640320.00 20484096.00Pool Paged Bytes 21152240.48 20766720.00 21159936.00Database Page Fault Stalls/sec0.0000.000 0.000Performance log C:\Jetstress\Backup_To_Disk_2006_9_7_1-29-27.blg is saved. 9/6/2006 10:57:30 PM -- Environment validation results: Detected JetstressUI version: 6.5.7830.0 Detected operating system: Microsoft Windows Server 2003 R2 (5.2.3790.0) Service Pack 1 Detected ESE.dll version: 6.5.7638.2 Detected ESEPERF.dll version: 6.5.7638.1 9/6/2006 10:57:30 PM -- Validating input parameters, it may take a few minutes... 9/6/2006 10:57:30 PM -- Attaching databases ... (it may take a few minutes if the databases are in dirty shutdown state) 9/6/2006 10:57:35 PM -- Loading performance counters... 9/6/2006 10:57:35 PM -- Instance5660.1: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Instance5660.2: IO parameters are thread (4), insert (17), replace (70), delete (5), and lazy commit (90) 9/6/2006 10:57:35 PM -- Start Jetstress test... 9/6/2006 10:57:36 PM -- Starting StreamingBackup test run... 9/6/2006 10:57:38 PM -- Performance logging started. 9/6/2006 10:57:38 PM -- Performance data will be saved to C:\Jetstress\Performance_LogGeneration_2006_9_6_22-57-35.blg. 9/7/2006 12:33:38 AM -- Adding new data to the performance log file... 9/7/2006 12:33:53 AM -- Performance logging stopped. 32 Untitled Document9/7/2006 12:33:53 AM -- Stopping Jetstress... 9/7/2006 12:34:52 AM -- Creating test report ... 9/7/2006 12:34:53 AM -- Volume F: has 0.0169 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume E: has 0.0167 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume L: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume L: has 0.0019 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Volume M: has 0.0001 for Avg. Disk sec/Write. 9/7/2006 12:34:53 AM -- Volume M: has 0.0013 for Avg. Disk sec/Read. 9/7/2006 12:34:53 AM -- Test has 0 Max Database Page Fault Stalls/sec. 9/7/2006 12:34:53 AM -- Test has 0 Database Page Fault Stalls/sec samples higher than 0. 9/7/2006 12:34:53 AM -- Soft recovery may take a while depending on the number of log files. 9/7/2006 12:34:55 AM -- Performance logging started. 9/7/2006 12:34:55 AM -- Performance data will be saved to C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.blg. 9/7/2006 12:51:11 AM -- Adding new data to the performance log file... 9/7/2006 12:51:14 AM -- Performance logging stopped. 9/7/2006 12:51:14 AM -- Soft recovery is completed, please open C:\Jetstress\SoftRecovery_2006_9_7_0-34-53.html for soft recovery result. 9/7/2006 12:51:16 AM -- Performance logging started. 9/7/2006 12:51:16 AM -- Performance data will be saved to C:\Jetstress\DatabaseChecksum_2006_9_7_0-51-14.blg. 9/7/2006 12:51:16 AM -- Checksum validation may take a while depending on the file sizes. 9/7/2006 1:28:03 AM -- Database checksum in progress: Storage Group #1 (100%), and Storage Group #2 (100%). 9/7/2006 1:28:03 AM -- Performance logging stopped. 9/7/2006 1:28:03 AM -- Checksum is completed, please open C:\Jetstress\DatabaseChecksum_2006_9_7_0-51-14.html for checksum result. 9/7/2006 1:28:05 AM -- Performance logging started. 9/7/2006 1:28:05 AM -- Performance data will be saved to C:\Jetstress\LogChecksum_2006_9_7_1-28-3.blg. 9/7/2006 1:28:05 AM -- Checksum validation may take a while depending on the file sizes. 9/7/2006 1:29:27 AM -- Log checksum in progress: (100 files(s) passed), and (100 files(s) passed). 9/7/2006 1:29:27 AM -- Performance logging stopped. 9/7/2006 1:29:27 AM -- Storage Groups being Backed up... The Backup Time may take hours depending on the size of the Storage Group. 9/7/2006 1:44:42 AM -- Performance logging started. 9/7/2006 1:44:42 AM -- Performance data will be saved to C:\Jetstress\Backup_To_Disk_2006_9_7_1-29-27.blg. 9/7/2006 1:44:42 AM -- Loading performance counters... 9/7/2006 3:56:47 AM -- Adding new data to the performance log file... 9/7/2006 3:56:48 AM -- Backup Complete! 9/7/2006 3:56:50 AM -- Performance logging stopped. 9/7/2006 3:56:50 AM -- Creating test report ... 33 Untitled DocumentFor more information For further information on HP Exchange solutions including best practices and additional testing of HP StorageWorks or ProLiant products with Exchange, please see http://www.hp.com/solutions/exchange For additional information on HP ProLiant Storage and StorageWorks products including the AiO600, please see, http://h18000.www1.hp.com/storage/arraysystems.htmlFor general information on Exchange sizing and best practices, there are several white papers and planning tools that are free to download from, http://www.hp.com/solutions/activeanswers 2006 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein. Intel, Celeron, and Xeon are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation. 4AA0-6976ENW, Rev 2, December 2006
