Guest

Unified Computing

Cisco UCS C210 M2 2500 Mailbox Resiliency Microsoft Exchange 2010 Storage Solution

  • Viewing Options

  • PDF (767.5 KB)
  • Feedback

White Paper

June 2012
Description: Image

Overview

This document provides information on Cisco Systems, Inc. storage solution for Microsoft Exchange Server, based on the Microsoft Exchange Solution Reviewed Program (ESRP)-Storage program*. For any questions or comments regarding the contents of this document, see Contact for Additional Information.
*The ESRP-Storage program was developed by Microsoft Corporation to provide a common storage testing framework for vendors to provide information on its storage solutions for Microsoft Exchange Server software. For more details on the Microsoft ESRP-Storage program, please go to http://www.microsoft.com/technet/prodtechnol/exchange/2007/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.
The information contained in this document represents the current view of Cisco Systems, Inc. on the issues discussed as of the date of publication. Due to changing market conditions, it should not be interpreted to be a commitment on the part of Cisco Systems, Inc., and Cisco Systems, Inc. cannot guarantee the accuracy of any information presented after the date of publication.

Cisco Unified Computing System

Today's data center managers are faced with the arduous task of reducing expenses while increasing responsiveness. Finding ways to better address rising data center capital and operations costs has companies pursuing strategies such as infrastructure consolidation and virtualization. Although this continues to offer dramatic increases in server utilization, virtualization alone is not the answer. In fact, the increase in virtual machines is driving the need for more software, networking and storage resources that are increasing costs and adding complexity within the data center. At the same time, companies are looking for architectures and solutions that will allow them to improve their service delivery capabilities. The industry is transitioning away from the rigid, inflexible platforms and moving toward more flexible, integrated, and virtualized environments. Cisco's Unified Computing System (UCS) was designed from the ground up to specifically meet these challenges.

Cisco UCS Solution

Cisco ® UCS C-Series Rack-Mount Servers extend unified computing innovations to an industry-standard form factor to help reduce total cost of ownership (TCO) and increase business agility. Designed to operate both in standalone environments and as part of the Cisco Unified Computing System , the series employs Cisco technology to help customers handle the most challenging workloads. The series incorporates a standards-based unified network fabric and Cisco VN-Link virtualization and protects customer investments with a future migration path to unified computing.
The Cisco UCS C210 M2 General-Purpose Rack-Mount Server is a two-socket, two-rack-unit (2RU) rack-mount server housing up to 16 internal small form-factor (SFF) SAS, SATA or SSD drives for a total of up to 16 terabytes (TB) of storage (Figure 1). The Cisco UCS C210 M2 server is designed to balance performance, density, and efficiency for workloads requiring economical, high-capacity, reliable, internal storage. Based on six-core Intel ® Xeon ® 5600 series processors, the server is built for applications including virtualization, network file servers and appliances, storage servers, database servers, and content-delivery servers.
Building on the success of the Cisco UCS C210 M1 General-Purpose Rack-Mount Server, the Cisco UCS C210 M2 server (Figure 1) extends the capabilities of the Cisco Unified Computing System with the next generation of Intel processor technology: Intel ® Xeon ® 5600 series processors. These powerful processors deliver more cores, threads, and cache, all within a similar power envelope, with even faster payback, greater productivity, and better energy efficiency than preceding models. When put into production, Cisco Unified Computing System and Intel Xeon 5600 series processors together offer further reductions in TCO, increased business agility, and another big leap forward in data center virtualization.

Figure 1. Cisco UCS C210 M2 Rack-Mount Server

The Cisco UCS C210 M2 server extends Cisco's product portfolio to meet the needs of customers that choose to deploy rack-mount servers. The server enables organizations to deploy systems incrementally, using as many or as few servers as needed, on a schedule that best meets the organization's timing and budget.
Designed to operate both in standalone environments and as part of the Cisco Unified Computing System, the server combines high-capacity disk storage and I/O configurations with Cisco innovations, including a unified network fabric and network-aware Cisco VN-Link technology.
The server brings differentiation and value to what has been a commodity market with products not optimized to meet the needs of virtualized data centers. Available from Cisco and its data center network infrastructure (DCNI) partners, the server advances the rack-mount server market with the features outlined in Table 1.

Table 1. Features and Benefits

Feature

Benefit

10-Gbps unified network fabric

• Low-latency, lossless, 10-Gbps Ethernet and industry-standard Fibre Channel over Ethernet (FCoE) fabric
• Wire-once deployment model in which changing I/O configurations no longer means installing adapters and recabling racks and switches
• Fewer interface cards, cables, and upstream network ports to purchase, power, configure, and maintain

Virtualization optimization

• Cisco VN-Link technology, I/O virtualization, and Intel Xeon 5600 series processor features, extending the network directly to virtual machines
• Consistent and scalable operational model
• Increased security and efficiency with reduced complexity

Unified management*
(when integrated into the Cisco Unified Computing System)

• Entire solution managed as a single entity with Cisco UCS Manager, improving operational efficiency and flexibility
• Service profiles and templates that implement role- and policy-based management, enabling more effective use of skilled server, network, and storage administrators
• Automated provisioning and increased business agility, allowing data center managers to provision applications in minutes rather than days

Six-core Intel Xeon 5600 series processors

• Intelligent performance that automatically adjusts processor performance to meet application demands, increasing performance when needed and achieving substantial energy savings when not
• Automated energy efficiency that reduces energy costs by automatically putting the processor and memory in the lowest available power state while still delivering the performance required
• Flexible virtualization technology that optimizes performance for virtualized environments, including processor support for migration and direct I/O
• With more cores, threads, and cache, in a similar power envelope, the Cisco Unified Computing System and Intel Xeon 5600 series processors together offer further reductions in TCO, increased business agility, and another big leap forward in data center virtualization.
• Cisco C-series servers keep pace with Intel Xeon processor innovation by offering the latest series 5600 processors with an increase in processor frequency and improved security features. With the increased clock speed, the Intel Xeon 5600 series based UCS C-Series rack mount servers will offer improved price/performance making UCS servers one of the best values in the industry.

High-capacity, flexible internal storage

• Up to 16 front-accessible, hot-swappable, SFF SAS, SATA or SSD drives for local storage, providing redundancy options and ease of serviceability
• Balanced performance and capacity to best meet application needs:
• 15,000 RPM SAS drives for highest performance
• 10,000 RPM SAS drives for high performance and value
• 7200-RPM SATA drives for high capacity and value

RAID 0, 1, 5, 6, 10, 50, and 60 support

A choice of RAID controllers to provide data protection for up to 16 SAS, SATA or SSD drives in PCIe and mezzanine card form factors

Cisco UCS Integrated Management Controller

• Web user interface for server management; remote keyboard, video, and mouse (KVM); virtual media; and administration
• Virtual media support for remote KVM and CD and DVD drives as if local
• Intelligent Platform Management Interface (IPMI) 2.0 support for out-of-band management through third-party enterprise management systems
• Command-line interface (CLI) for server management

Fast-memory support

12 DIMM slots supporting up to 192 GB of 1333-MHz memory for optimal performance

Redundant fans and power supplies

Dual-redundant fans and power supplies for enterprise-class reliability and uptime

5 PCI Express (PCIe) 2.0 slots

• Flexibility, increased performance, and compatibility with industry standards
• PCIe 2.0 slots, which double bandwidth over the previous generation and offer more flexibility while maintaining compatibility with PCIe 1
• I/O performance and flexibility with 2 full-height, full-length and 3 full-height half-length x8 PCIe slots, all with x16 connectors

Integrated dual-port Gigabit Ethernet

• Outstanding network I/O performance and increased network efficiency and flexibility.
• Increased network availability when configured in failover configurations

Optional optical drive

Direct front-panel read/write access to CD and DVD media

Solution Description

The tested solution (Figure 2) provides a highly available Exchange 2010 deployment for 2500 mailbox users. Each Exchange server runs the combined Mailbox, Client Access Server, and Hub Transport roles on Cisco C210 M2 rack mount servers. The Cisco C210 M2 rack mount servers are efficient, occupying only 2 RU of rack space. The primary datacenter has two Exchange servers in a Data Availability Group. The Data Availability Group uses continuous replication to replicate mailbox database data between member servers. During normal operation each Exchange server runs 1250 active mailboxes.
If one of the two Exchange servers in the primary datacenter needs to be taken offline or fails, the surviving Exchange server is capable of servicing all 2500 mailboxes without noticeable service degradation. The Client Access Server and Hub Transport server roles, which run on each exchange server, are also redundant and are capable of sustaining the entire workload during peak work hours without noticeable service degradation.

Figure 2. Solution Architecture

The test was conducted on a C210 M2 Cisco UCS Server, firmware version 1.4 (1a), Bios Version: C200 1.4.1.0 (BuildDate 07/13/2011). The server is powered by two (2) 6-Core 12 thread Xeon E5649@2.53Ghz CPUs and contains 48Gb (6 x 8192MB Synchronous DIMMs) 1333mhz RAM. The storage adaptor used to create the RAID 1, and RAID 10 Storage Volumes is an LSI MegaRAID SAS 9261-8i with 512 MB Cache.

Server Platform

Cisco UCS C-Series 210 M2 Server

Firmware

1.4 (1a)

Bios

C200 1.4.1.0(BuildDate 07/13/2011)

RAM

48Gb(6 x 8192MB Synchronous DIMMs) 1333mhz RAM

RAID Adapter

LSI MegaRAID SAS 9261-8i 512MB Cache

CPU

2 6-Core 12 thread Xeon

 

E5649@2.53Ghz

Storage

16 Seagate 1TB 7.2K RPM SATA ST91000640NS 2.5" 1 TB Drives

Figure 3. Cisco UCS C210 M1/M2 Front and Back View

The two drive RAID 1 solution for the system drive is illustrated in Figure 3. RAID 1 was created using the RAID controller's WebBios screen which can be accessed during the boot up process using the Ctrl+H keystroke.

Figure 4. RAID 1 Solution

For additional information regarding the RAID Controller configuration and Cisco UCS C210 configuration options, please refer to:
The 14 drive RAID 10 solution for the data drive is illustrated in Figure 4. The drives are configured within the controller as 7 spans of mirrors (RAID 1) with a stripe throughout the spans (RAID 0). The result is RAID 10 (1+0) redundancy for the drive that houses the database.

Figure 5. RAID 10 Solution

The following links will aid in visualizing the provided solution:

http://www.cisco.com/en/US/prod/collateral/ps10265/ps10493/data_sheet_c78-587522.html

http://www.cisco.com/en/US/prod/ps10265/ps10493/ps10516/ucs210_kaon_model_preso.html

http://www.cisco.com/en/US/products/ps10889/index.html

The ESRP-Storage program focuses on storage solution testing to address performance and reliability issues with storage design. However, storage is not the only factor to take into consideration when designing a scale up Exchange solution. Other factors that affect the server scalability are:

• Server processor utilization

• Server physical and virtual memory limitations

• Resource requirements for other applications

• Directory and network service latencies

• Network infrastructure limitations

• Replication and recovery requirements

• 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
http://technet.microsoft.com/en-us/library/dd335215.aspx.

Targeted Customer Profile

This solution is intended for small-to-medium sized organizations:

• 2500 Mailboxes

• One Host attached to storage system

• Mailbox Resiliency = Data Availability Group (DAG)

• Three Database Copies Per Server

• Mailbox quota: 1352 MB per mailbox

• User IO profile: 0.18 IOPS per mailbox

• 24x7 Background Database Maintenance utilized

• Eight Databases per server

Tested Deployment

The following tables summarize the testing environment:

Table 2. Simulated Exchange Configuration

Number of Exchange mailboxes simulated

2500

Number of Database Availability Groups (DAGs)

1

Number of servers/DAG

1

Number of active mailboxes/server

2500

Number of databases/host

8

Number of copies/database

3

Number of mailboxes/database

312

Simulated profile: I/O's per second per mailbox (IOPS, include 20 percent headroom)

.18

Database LUN size

6512.61 GB

Log LUN size

930gb

Total database size for performance testing

3296gb

% storage capacity used by Exchange database**

44.29%=(3296gb/7442gb)

**Storage performance characteristics change based on the percentage utilization of the individual disks. Tests that use a small percentage of the storage (~25%) may exhibit reduced throughput if the storage capacity utilization is significantly increased beyond what is tested in this paper.

Table 3. Storage Hardware

Storage Connectivity (Fiber Channel, SAS, SATA, iSCSI)

SATA

Storage model and OS/firmware revision

LSI MegaRAID SAS 9261-8i

Firmware Version: 2.12-.133-1322

Storage cache

397MB Cache Memory Size

Number of storage controllers

1

Number of storage ports

8

Maximum bandwidth of storage connectivity to host

N/A

Switch type/model/firmware revision

N/A

HBA model and firmware

LSI 9261-8i MegaRAID SAS HBA Firmware Version: 2.12-.133-1322

Number of HBA's/host

N/A

Host server type

Two (2) 6-Core 12 thread Xeon E5649@2.53Ghz CPUs and contains 48Gb (12 x 8192MB Synchronous DIMMs) 1067mhz RAM

Total number of disks tested in solution

16

Maximum number of spindles can be hosted in the storage

16

Table 4. Storage Software

HBA driver

MEGASAS Storport Driver for Windows 7\Server 2008 R2 for x64 megasas2.sys 4.32.0.64

HBA QueueTarget Setting

N/A

HBA QueueDepth Setting

N/A

Multi-Pathing

N/A

Host OS

Windows 2008R2 Enterprise SP1

ESE.dll file version

14.1.218.12

Replication solution name/version

N/A

Table 5. Storage Disk Configuration (Mailbox Store Disks)

Disk type, speed and firmware revision

Seagate 1TB 7.2K RPM SATA ST91000640NS 2.5" 1 TB Drives

Raw capacity per disk (GB)

931.513

Number of physical disks in test

14

Total raw storage capacity (GB)

13041.182

Disk slice size (GB)

N/A

Number of slices per LUN or number of disks per LUN

N/A

RAID level

RAID 10

Total formatted capacity

6512.61 GB

Storage capacity utilization

6512.61/13041.182=49.93%

Database capacity utilization

4152/13041.182=31.84%

Table 6. Storage Disk Configuration (Transactional Log Disks)

Disk type, speed and firmware revision

Seagate 1TB 7.2K RPM SATA ST91000640NS 2.5" 1 TB Drives

Raw capacity per disk (GB)

931.513

Number of Spindles in test

1

Total raw storage capacity (GB)

931.513

Disk slice size (GB)

N/A

Number of slices per LUN or number of disks per LUN

N/A

RAID level

RAID 1

Total formatted capacity

930.29 GB

Best Practices

Set up the system using the following table as a guide.

Table 7. Guide for Best Practices

Drive Type

Drive Letter

Raid Type

Format

System Drive

Drive C:

Raid 1

Contains Operating System and Log Files; format as NTFS; 64K disk allocation size

Data Drive

Drive F:

Raid 10

Format as GPT drive. Contains Database; 64K disk allocation size

The outlined design fulfills the capacity and performance needs for 2500 users, each with a Mailbox limit of 1352 MB. The user profile of 0.18 IOPS per mailbox includes a margin for 20 percent overhead.
Exchange server 2010 is a disk I/O intensive application. Tests were conducted using the Microsoft tool Jetstress 2010, 14.01.0225.017.
Install current firmware and device drivers for all devices in the server, such as network adapters, even if the device will not be used in the configuration. Current Cisco C210 M2 firmware and device drivers for Windows can be found on the UCS Server Firmware ISO and the UCS Drivers ISO. Both can be downloaded from www.cisco.com.
The following practices will help improve the I/O subsystem performance:

• Exchange 2010 is an IO intensive application. Do not share Exchange 2010 storage resources with other applications as performance may be negatively impacted.

• Store the Exchange databases and transaction logs on separate NTFS drive volumes. The Exchange Server 2010 resiliency solution no longer requires separate spindles for log and database files. However, placing the databases on a dedicated NTFS drive and the transaction logs on the system drive allows more space for Exchange databases and enables the deployment of larger mailboxes.

• Windows NTFS allocation unit size for Exchange 2010 database partitions should be set to 64K disk allocation size.

• Exchange Server 2010 storage latencies are related to the number of disks available for a given workload. Windows Performance Monitor may be used to monitor Exchange Server 2010 database counters. Average database read latencies (Avg. Disk sec/Read) should not exceed 20ms.

• Choose Enterprise class SATA disks for storage considerations as storage/design requirements indicate high capacity, moderate performance, and moderate power utilization and SATA storage has very good heat, vibration, and reliability characteristics. The disks used in this test are Seagate Constellation ST91000640N 2.5 inch, 7200 RPM, 1TB SATA drives.

• Windows Server 2008 R2 does not require the use of diskpart.exe to define sector boundary alignment as Window Server 2003 did since the sector boundaries are created automatically. Logical volumes are created using the Windows logical disk manager during the setup phase of the OS installation. Also, using Windows 2008R2 setup, the disk partitions are created with an automatic 1 MB offset (except when the disk capacity is less than 4 GB).

• Format the Database RAID 10 Array as a GPT disk if it is larger than 2TB. In this case, the 14 1TB RAID 10 disks yield approximately 7TB, and a GPT format is recommended.

RAID Controller best practices are to create a 256K strip size on the arrays and then enable read ahead, writeback, direct IO, Drive Cache disable, and Disable BGI parameters. Refer to the following table:

Table 8. RAID Controller Best Practices

Parameter

Value

Strip Size

256

Access Policy

RW

Read Policy

Ahead

Write Policy

WBack

IO Policy

Direct

Drive Cache

Disable

Disable BGI

No

Create the System Drive so that it will contain Windows 2008 R2 SP1 Enterprise using RAID 1 (2 spindles) and the Data Drive holding the Jetstress database using RAID 10 (14 spindles). Point the WebBrowser to the UCS C210 Server by entering its IP address and within the Cisco UCS Console, open the KVM, exercise the Ctrl+Alt+Del Macro from the file menu and while the host computer is booting press CTRL+H at the following screen in order to enter the RAID bios menu:

• Copyright© LSI Corporation

• Press <Ctrl><H> for WebBIOS

A message will then indicate that WebBios will load after post completes.
In the next screen, LSI WebBios, you can set up the RAID Controller drive.
Further RAID adjustments can be made within the OS after installing the LSI Megaraid Utility, now v.9, in Windows 2008 R2 SP1.
For Exchange Server 2010 Mailbox Storage Design, please visit
http://technet.microsoft.com/en-us/library/dd346703.aspx.

Backup Strategy

The tests performed for backup are: backup-to-disk (read only) and log replay. The backup-to-disk test measures the read I/O performance by running a checksum on all the databases and log files. This test can help determine the level of database read throughput that can be achieved during backups. The backup speed and throughput achieved will depend upon the backup device used. 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 the restore times and the database write throughput that can be achieved during a log recovery.

Contact For Additional Information

To find out more about Cisco Unified Computing, please visit: http://www.cisco.com/go/ucs.
To learn more about running Microsoft Enterprise Applications on Cisco Unified Computing System, please visit: http://www.cisco.com/go/microsoft.

Test Results Summary

This section provides a high level summary of the test data from ESRP and the link to the detailed html reports which are generated by ESRP testing framework. Please click the underlined headings below to view the HTML report for each test.

Reliability

A number of tests in the framework will check 24 hour Reliability tests runs. The goal is to verify that the storage can handle high IO load for a long period of time. Both log and database files were analyzed for integrity after the stress test to help ensure that there was no database/log corruption.
The following list provides an overview:

• Any errors reported in the saved eventlog file? No

• Any errors reported during the database and log checksum process? No

• If backup to disk test is done, any errors reported during the process? No

• Any errors during database checksum on the remote storage database? N/A

Storage Performance Results

The Primary Storage performance test is designed to exercise the storage simulating maximum sustainable Exchange type of IO for two hours. The test measures the length of time needed for the storage subsystem to respond to an IO under load. The data below is the sum of all of the logical disk I/O's and average of all the logical disks I/O latency in the two hours test duration. Each server is listed separately and the aggregate numbers across all servers are listed as well

Individual Performance Across All Servers Metrics

The sum of I/O's across all servers in the solution including their average latency.

Database I/O

Database Disks Transfers/sec

651.699

Database Disks Reads/sec

370.981

Database Disks Writes/sec

280.717

Average Database Disk Read Latency (ms)

14.580

Average Database Disk Write Latency (ms)

1.342

Transaction Log I/O

Log Disks Writes/sec

248.284

Average Log Disk Write Latency (ms)

1.065

Database Backup/Recover Performance

There are two tests reports in this section. The first one measures the sequential read rate of the database files, and the second measures the recovery/replay performance (playing transaction logs in to the database).

Database Read-Only Performance

This test measures the maximum rate at which databases can be backed up through VSS. The following table shows the average rate for a single database file.

MB read/sec per database

30.846

MB read/sec total per server

246.77

Transaction Log Recovery/Replay Performance

This test measures the maximum rate at which the log files can be played against the databases. The following table shows the average rate for 500 log files played in a single storage group. Each log file is 1 MB in size.

Average time to play one Log file (sec)

2.8426

Conclusion

Our tests, outlined in this document, illustrate the fact that the Cisco UCS C210 M2 system is an excellent platform for provisioning an Exchange Resiliency 2010 Storage Solution. The tests clearly show that 2500 Exchange users have ample resources within a Cisco UCS system containing:

• Dual Xeon 6-core 12 thread 5649@2.53Ghz CPUs

• 48Gb (12 x 8192MB Synchronous DIMMs) 1067mhz RAM

• LSI 9261-8i MegaRAID SAS Raid Adapter

• 16 Seagate 1TB 7.2K RPM ST91000640NS 2.5" 1 TB SATA storage drives

To maintain the IOPs required by a Microsoft Exchange Resiliency solution as well as needed reserve to support continuous and uninterrupted use.
This document was developed by Cisco and reviewed by Microsoft Exchange Product team. The test results/data presented in this document are based on the tests introduced in the ESRP test framework. Customer should not quote the data directly for their 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 get the maximum throughput for a given solution. Instead, ESRP is focused on producing recommendations from vendors for Exchange application. Therefore, the data presented in this document should not be used for direct comparisons among the solutions.

Appendix A: Stress Testing

Stress Test Result Report

Test Summary

Overall Test Result

Pass

Machine Name

EXCHANGE1

Test Description

Create New Config 3 threads, .18 iops, 1352mb mailbox, 2500 users 48gb RAM Test

Test Start Time

5/29/2012 10:12:53 AM

Test End Time

5/30/2012 10:17:44 AM

Collection Start Time

5/29/2012 10:17:24 AM

Collection End Time

5/30/2012 10:17:21 AM

Jetstress Version

14.01.0225.017

ESE Version

14.02.0247.001

Operating System

Windows Server 2008 R2 Enterprise Service Pack 1 (6.1.7601.65536)

Performance Log

C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.blg

Database Sizing and Throughput

Achieved Transactional I/O per Second

613.269

Target Transactional I/O per Second

450

Initial Database Size (bytes)

3568849911808

Final Database Size (bytes)

3594108010496

Database Files (Count)

8

Jetstress System Parameters

Thread Count

3 (per database)

Minimum Database Cache

256.0 MB

Maximum Database Cache

2048.0 MB

Insert Operations

40%

Delete Operations

20%

Replace Operations

5%

Read Operations

35%

Lazy Commits

70%

Run Background Database Maintenance

True

Number of Copies per Database

3

Database Configuration

Instance2536.1

Log path: c:\log1

Database: F:\db1\Jetstress001001.edb

Instance2536.2

Log path: c:\log2

Database: F:\db2\Jetstress002001.edb

Instance2536.3

Log path: c:\log3

Database: F:\db3\Jetstress003001.edb

Instance2536.4

Log path: c:\log4

Database: F:\db4\Jetstress004001.edb

Instance2536.5

Log path: c:\log5

Database: F:\db5\Jetstress005001.edb

Instance2536.6

Log path: c:\log6

Database: F:\db6\Jetstress006001.edb

Instance2536.7

Log path: c:\log7

Database: F:\db7\Jetstress007001.edb

Instance2536.8

Log path: c:\log8

Database: f:\db8\Jetstress008001.edb

Transactional I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance2536.1

15.332

1.335

43.401

33.224

35274.674

35563.666

Instance2536.2

14.808

1.354

43.517

33.340

35250.679

35564.863

Instance2536.3

15.372

1.351

43.554

33.377

35319.407

35560.309

Instance2536.4

16.297

1.360

43.145

33.000

35277.267

35555.155

Instance2536.5

17.193

1.366

43.532

33.370

35131.347

35550.476

Instance2536.6

16.446

1.360

43.340

33.188

35278.988

35552.031

Instance2536.7

15.340

1.380

43.384

33.223

35267.686

35567.151

Instance2536.8

14.896

1.381

43.426

33.246

35315.183

35541.813

Transactional I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance2536.1

0.000

1.139

0.000

28.151

0.000

4633.049

Instance2536.2

0.000

1.145

0.000

28.319

0.000

4645.742

Instance2536.3

0.000

1.158

0.000

28.297

0.000

4643.700

Instance2536.4

0.000

1.170

0.000

28.091

0.000

4636.706

Instance2536.5

0.000

1.173

0.000

28.331

0.000

4631.251

Instance2536.6

0.000

1.171

0.000

28.204

0.000

4639.940

Instance2536.7

0.000

1.159

0.000

28.143

0.000

4645.343

Instance2536.8

0.000

1.157

0.000

28.161

0.000

4644.227

Background Database Maintenance I/O Performance

MSExchange Database ==> Instances

Database Maintenance IO Reads/sec

Database Maintenance IO Reads Average Bytes

Instance2536.1

26.790

261170.891

Instance2536.2

26.902

261170.996

Instance2536.3

26.362

261170.864

Instance2536.4

25.858

261145.761

Instance2536.5

24.900

261148.443

Instance2536.6

25.457

261179.648

Instance2536.7

26.391

261163.257

Instance2536.8

26.935

261131.667

Log Replication I/O Performance

MSExchange Database ==> Instances

I/O Log Reads/sec

I/O Log Reads Average Bytes

Instance2536.1

1.054

198551.356

Instance2536.2

1.063

201209.482

Instance2536.3

1.062

199583.992

Instance2536.4

1.054

198786.867

Instance2536.5

1.060

200033.054

Instance2536.6

1.057

200490.229

Instance2536.7

1.057

199925.927

Instance2536.8

1.057

199573.541

Total I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance2536.1

15.332

1.335

70.191

33.224

121493.909

35563.666

Instance2536.2

14.808

1.354

70.420

33.340

121558.828

35564.863

Instance2536.3

15.372

1.351

69.916

33.377

120477.676

35560.309

Instance2536.4

16.297

1.360

69.003

33.000

119917.675

35555.155

Instance2536.5

17.193

1.366

68.433

33.370

117371.699

35550.476

Instance2536.6

16.446

1.360

68.797

33.188

118868.735

35552.031

Instance2536.7

15.340

1.380

69.774

33.223

120707.702

35567.151

Instance2536.8

14.896

1.381

70.361

33.246

121759.630

35541.813

Total I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance2536.1

0.827

1.139

1.054

28.151

198551.356

4633.049

Instance2536.2

0.833

1.145

1.063

28.319

201209.482

4645.742

Instance2536.3

0.826

1.158

1.062

28.297

199583.992

4643.700

Instance2536.4

0.828

1.170

1.054

28.091

198786.867

4636.706

Instance2536.5

0.835

1.173

1.060

28.331

200033.054

4631.251

Instance2536.6

0.834

1.171

1.057

28.204

200490.229

4639.940

Instance2536.7

0.829

1.159

1.057

28.143

199925.927

4645.343

Instance2536.8

0.830

1.157

1.057

28.161

199573.541

4644.227

Host System Performance

Counter

Average

Minimum

Maximum

% Processor Time

1.081

0.000

8.249

Available MBytes

45116.092

45072.000

45310.000

Free System Page Table Entries

33555545.138

33555538.000

33557578.000

Transition Pages RePurposed/sec

0.000

0.000

0.000

Pool Nonpaged Bytes

87799566.220

87748608.000

87986176.000

Pool Paged Bytes

113182492.851

112648192.000

116183040.000

Database Page Fault Stalls/sec

0.000

0.000

0.000

Test Log

5/29/2012 10:12:53 AM -- Jetstress testing begins ...

5/29/2012 10:12:53 AM -- Preparing for testing ...

5/29/2012 10:13:02 AM -- Attaching databases ...

5/29/2012 10:13:02 AM -- Preparations for testing are complete.

5/29/2012 10:13:03 AM -- Starting transaction dispatch ..

5/29/2012 10:13:03 AM -- Database cache settings: (minimum: 256.0 MB, maximum: 2.0 GB)

5/29/2012 10:13:03 AM -- Database flush thresholds: (start: 20.5 MB, stop: 40.9 MB)

5/29/2012 10:13:11 AM -- Database read latency thresholds: (average: 20 msec/read, maximum: 100 msec/read).

5/29/2012 10:13:11 AM -- Log write latency thresholds: (average: 10 msec/write, maximum: 100 msec/write).

5/29/2012 10:13:21 AM -- Operation mix: Sessions 3, Inserts 40%, Deletes 20%, Replaces 5%, Reads 35%, Lazy Commits 70%.

5/29/2012 10:13:21 AM -- Performance logging started (interval: 15000 ms).

5/29/2012 10:13:21 AM -- Attaining prerequisites:

5/29/2012 10:17:24 AM -- \MSExchange Database(JetstressWin)\Database Cache Size, Last: 1941848000.0 (lower bound: 1932735000.0, upper bound: none)

5/30/2012 10:17:24 AM -- Performance logging has ended.

5/30/2012 10:17:24 AM -- JetInterop batch transaction stats: 222502, 223361, 223068, 222069, 223510, 222601, 222702 and 222837.

5/30/2012 10:17:24 AM -- Dispatching transactions ends.

5/30/2012 10:17:24 AM -- Shutting down databases ...

5/30/2012 10:17:44 AM -- Instance2536.1 (complete), Instance2536.2 (complete), Instance2536.3 (complete), Instance2536.4 (complete), Instance2536.5 (complete), Instance2536.6 (complete), Instance2536.7 (complete) and Instance2536.8 (complete)

5/30/2012 10:17:44 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.blg has 5614 samples.

5/30/2012 10:17:44 AM -- Creating test report ...

5/30/2012 10:18:59 AM -- Instance2536.1 has 15.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.1 has 1.1 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.1 has 1.1 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 14.8 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 1.1 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 1.1 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 15.4 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 16.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 17.2 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 16.4 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 15.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 14.9 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Test has 0 Maximum Database Page Fault Stalls/sec.

5/30/2012 10:18:59 AM -- The test has 0 Database Page Fault Stalls/sec samples higher than 0.

5/30/2012 10:18:59 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.xml has 5598 samples queried.

Appendix B: Performance Testing

Performance Test Result Report

Test Summary

Overall Test Result

Pass

Machine Name

EXCHANGE1

Test Description

Create New Config 3 threads, .18 iops, 1352mb mailbox, 2500 users 48gb RAM Test

Test Start Time

5/16/2012 9:37:03 AM

Test End Time

5/16/2012 11:41:44 AM

Collection Start Time

5/16/2012 9:41:25 AM

Collection End Time

5/16/2012 11:41:24 AM

Jetstress Version

14.01.0225.017

ESE Version

14.02.0247.001

Operating System

Windows Server 2008 R2 Enterprise Service Pack 1 (6.1.7601.65536)

Performance Log

C:\Program Files\Exchange Jetstress\Performance_2012_5_16_9_37_21.blg

Database Sizing and Throughput

Achieved Transactional I/O per Second

651.699

Target Transactional I/O per Second

450

Initial Database Size (bytes)

3553834303488

Final Database Size (bytes)

3556266999808

Database Files (Count)

8

Jetstress System Parameters

Thread Count

3 (per database)

Minimum Database Cache

256.0 MB

Maximum Database Cache

2048.0 MB

Insert Operations

40%

Delete Operations

20%

Replace Operations

5%

Read Operations

35%

Lazy Commits

70%

Run Background Database Maintenance

True

Number of Copies per Database

3

Database Configuration

Instance2536.1

Log path: c:\log1

Database: F:\db1\Jetstress001001.edb

Instance2536.2

Log path: c:\log2

Database: F:\db2\Jetstress002001.edb

Instance2536.3

Log path: c:\log3

Database: F:\db3\Jetstress003001.edb

Instance2536.4

Log path: c:\log4

Database: F:\db4\Jetstress004001.edb

Instance2536.5

Log path: c:\log5

Database: F:\db5\Jetstress005001.edb

Instance2536.6

Log path: c:\log6

Database: F:\db6\Jetstress006001.edb

Instance2536.7

Log path: c:\log7

Database: F:\db7\Jetstress007001.edb

Instance2536.8

Log path: c:\log8

Database: f:\db8\Jetstress008001.edb

Total I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance3852.1

14.115

1.324

46.309

35.036

34901.856

36404.931

Instance3852.2

13.560

1.294

46.580

35.266

35037.968

36378.305

Instance3852.3

16.008

1.353

46.071

34.822

34907.609

36432.698

Instance3852.4

14.085

1.311

46.918

35.727

34781.188

36460.595

Instance3852.5

15.197

1.327

46.197

34.801

34916.010

36451.800

Instance3852.6

15.273

1.354

46.443

35.194

34927.510

36455.343

Instance3852.7

14.281

1.388

46.124

34.889

34811.357

36423.794

Instance3852.8

14.122

1.381

46.339

34.982

34725.054

36401.212

Total I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance3852.1

0.000

1.036

0.000

30.772

0.000

4633.090

Instance3852.2

0.000

1.046

0.000

30.866

0.000

4648.594

Instance3852.3

0.000

1.089

0.000

31.104

0.000

4653.530

Instance3852.4

0.000

1.069

0.000

31.573

0.000

4664.550

Instance3852.5

0.000

1.075

0.000

30.714

0.000

4657.271

Instance3852.6

0.000

1.073

0.000

31.444

0.000

4670.224

Instance3852.7

0.000

1.070

0.000

30.946

0.000

4687.627

Instance3852.8

0.000

1.065

0.000

30.865

0.000

4677.848

Background Database Maintenance I/O Performance

MSExchange Database ==> Instances

Database Maintenance IO Reads/sec

Database Maintenance IO Reads Average Bytes

Instance3852.1

27.918

261158.343

Instance3852.2

28.148

261290.040

Instance3852.3

25.863

261297.587

Instance3852.4

27.723

261247.570

Instance3852.5

26.867

261210.519

Instance3852.6

26.489

261125.447

Instance3852.7

27.456

261242.280

Instance3852.8

27.836

261262.051

Log Replication I/O Performance

MSExchange Database ==> Instances

I/O Log Reads/sec

I/O Log Reads Average Bytes

Instance3852.1

1.155

209958.581

Instance3852.2

1.160

207943.076

Instance3852.3

1.170

215483.807

Instance3852.4

1.190

214981.514

Instance3852.5

1.160

214516.361

Instance3852.6

1.183

217492.980

Instance3852.7

1.173

213976.927

Instance3852.8

1.170

211967.754

Total I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance3852.1

14.115

1.324

74.227

35.036

119999.774

36404.931

Instance3852.2

13.560

1.294

74.729

35.266

120260.930

36378.305

Instance3852.3

16.008

1.353

71.934

34.822

116304.536

36432.698

Instance3852.4

14.085

1.311

74.641

35.727

118893.659

36460.595

Instance3852.5

15.197

1.327

73.064

34.801

118128.321

36451.800

Instance3852.6

15.273

1.354

72.932

35.194

117082.310

36455.343

Instance3852.7

14.281

1.388

73.579

34.889

119302.007

36423.794

Instance3852.8

14.122

1.381

74.175

34.982

119737.695

36401.212

Total I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance3852.1

0.845

1.036

1.155

30.772

209958.581

4633.090

Instance3852.2

0.814

1.046

1.160

30.866

207943.076

4648.594

Instance3852.3

0.872

1.089

1.170

31.104

215483.807

4653.530

Instance3852.4

0.850

1.069

1.190

31.573

214981.514

4664.550

Instance3852.5

0.915

1.075

1.160

30.714

214516.361

4657.271

Instance3852.6

0.845

1.073

1.183

31.444

217492.980

4670.224

Instance3852.7

0.895

1.070

1.173

30.946

213976.927

4687.627

Instance3852.8

0.812

1.065

1.170

30.865

211967.754

4677.848

Host System Performance

Counter

Average

Minimum

Maximum

% Processor Time

1.121

0.000

4.660

Available MBytes

45211.280

45206.000

45323.000

Free System Page Table Entries

33555544.375

33555542.000

33555546.000

Transition Pages RePurposed/sec

0.000

0.000

0.000

Pool Nonpaged Bytes

87109031.724

87093248.000

87179264.000

Pool Paged Bytes

125085113.379

125018112.000

125165568.000

Database Page Fault Stalls/sec

0.000

0.000

0.000

Test Log

5/16/2012 9:37:03 AM -- Jetstress testing begins ...

5/16/2012 9:37:03 AM -- Preparing for testing ...

5/16/2012 9:37:12 AM -- Attaching databases ...

5/16/2012 9:37:12 AM -- Preparations for testing are complete.

5/16/2012 9:37:12 AM -- Starting transaction dispatch ..

5/16/2012 9:37:12 AM -- Database cache settings: (minimum: 256.0 MB, maximum: 2.0 GB)

5/16/2012 9:37:12 AM -- Database flush thresholds: (start: 20.5 MB, stop: 40.9 MB)

5/16/2012 9:37:21 AM -- Database read latency thresholds: (average: 20 msec/read, maximum: 100 msec/read).

5/16/2012 9:37:21 AM -- Log write latency thresholds: (average: 10 msec/write, maximum: 100 msec/write).

5/16/2012 9:37:32 AM -- Operation mix: Sessions 3, Inserts 40%, Deletes 20%, Replaces 5%, Reads 35%, Lazy Commits 70%.

5/16/2012 9:37:32 AM -- Performance logging started (interval: 15000 ms).

5/16/2012 9:37:32 AM -- Attaining prerequisites:

5/16/2012 9:41:25 AM -- \MSExchange Database(JetstressWin)\Database Cache Size, Last: 1933160000.0 (lower bound: 1932735000.0, upper bound: none)

5/16/2012 11:41:25 AM -- Performance logging has ended.

5/16/2012 11:41:25 AM -- JetInterop batch transaction stats: 20975, 20966, 20985, 21149, 20968, 21169, 20983 and 20896.

5/16/2012 11:41:26 AM -- Dispatching transactions ends.

5/16/2012 11:41:26 AM -- Shutting down databases ...

5/16/2012 11:41:44 AM -- Instance3852.1 (complete), Instance3852.2 (complete), Instance3852.3 (complete), Instance3852.4 (complete), Instance3852.5 (complete), Instance3852.6 (complete), Instance3852.7 (complete) and Instance3852.8 (complete)

5/16/2012 11:41:44 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_16_9_37_21.blg has 479 samples.

5/16/2012 11:41:44 AM -- Creating test report ...

5/16/2012 11:41:51 AM -- Instance3852.1 has 14.1 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.1 has 1.0 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.1 has 1.0 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.2 has 13.6 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.2 has 1.0 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.2 has 1.0 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.3 has 16.0 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.3 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.3 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.4 has 14.1 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.4 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.4 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.5 has 15.2 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.5 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.5 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.6 has 15.3 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.6 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.6 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.7 has 14.3 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.7 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.7 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.8 has 14.1 for I/O Database Reads Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.8 has 1.1 for I/O Log Writes Average Latency.

5/16/2012 11:41:51 AM -- Instance3852.8 has 1.1 for I/O Log Reads Average Latency.

5/16/2012 11:41:51 AM -- Test has 0 Maximum Database Page Fault Stalls/sec.

5/16/2012 11:41:51 AM -- The test has 0 Database Page Fault Stalls/sec samples higher than 0.

5/16/2012 11:41:51 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_16_9_37_21.xml has 463 samples queried.

Appendix C: Streaming Backup Testing

Database backup Test Result Report
Database Backup Statistics - All

Database Instance

Database Size (MBytes)

Elapsed Backup Time

MBytes Transferred/sec

Instance2804.1

423377.09

03:46:20

31.17

Instance2804.2

423385.09

03:49:54

30.69

Instance2804.3

423377.09

04:05:46

28.71

Instance2804.4

423393.09

04:06:04

28.68

Instance2804.5

423385.09

03:58:50

29.54

Instance2804.6

423377.09

03:39:08

32.20

Instance2804.7

423369.09

03:25:07

34.40

Instance2804.8

423369.09

03:44:52

31.38

Jetstress System Parameters

Thread Count

3 (per database)

Minimum Database Cache

256.0 MB

Maximum Database Cache

2048.0 MB

Insert Operations

40%

Delete Operations

20%

Replace Operations

5%

Read Operations

35%

Lazy Commits

70%

Database Configuration

Instance2804.1

Log path: c:\log1

Backup path: f:\bak1

Database: F:\db1\Jetstress001001.edb

Instance2804.2

Log path: c:\log2

Backup path: f:\bak2

Database: F:\db2\Jetstress002001.edb

Instance2804.3

Log path: c:\log3

Backup path: f:\bak3

Database: F:\db3\Jetstress003001.edb

Instance2804.4

Log path: c:\log4

Backup path: f:\bak4

Database: F:\db4\Jetstress004001.edb

Instance2804.5

Log path: c:\log5

Backup path: f:\bak5

Database: F:\db5\Jetstress005001.edb

Instance2804.6

Log path: c:\log6

Backup path: f:\bak6

Database: F:\db6\Jetstress006001.edb

Instance2804.7

Log path: c:\log7

Backup path: f:\bak7

Database: F:\db7\Jetstress007001.edb

Instance2804.8

Log path: c:\log8

Backup path: c:\bak8

Database: f:\db8\Jetstress008001.edb

Transactional I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance2804.1

8.826

0.000

124.416

0.000

262144.000

0.000

Instance2804.2

13.834

0.000

122.521

0.000

262144.000

0.000

Instance2804.3

17.156

0.000

114.459

0.000

262144.000

0.000

Instance2804.4

16.407

0.000

114.141

0.000

262144.000

0.000

Instance2804.5

14.954

0.000

117.967

0.000

262144.000

0.000

Instance2804.6

9.791

0.000

128.681

0.000

262144.000

0.000

Instance2804.7

9.078

0.000

137.536

0.000

262144.000

0.000

Instance2804.8

9.867

0.000

125.204

0.000

262144.000

0.000

Transactional I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance2804.1

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.2

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.3

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.4

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.5

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.6

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.7

0.000

0.000

0.000

0.000

0.000

0.000

Instance2804.8

0.000

0.000

0.000

0.000

0.000

0.000

Host System Performance

Counter

Average

Minimum

Maximum

% Processor Time

1.940

0.532

3.455

Available MBytes

40716.936

37226.000

43721.000

Free System Page Table Entries

33555790.870

33555534.000

33557079.000

Transition Pages RePurposed/sec

58113.461

0.000

70162.089

Pool Nonpaged Bytes

94466031.039

54333440.000

128966656.000

Pool Paged Bytes

3405124457.474

141398016.000

6215434240.000

Database Page Fault Stalls/sec

0.000

0.000

0.000

Test Log

5/11/2012 10:19:12 AM -- Jetstress testing begins ...

5/11/2012 10:19:12 AM -- Preparing for testing ...

5/11/2012 10:19:21 AM -- Attaching databases ...

5/11/2012 10:19:21 AM -- Preparations for testing are complete.

5/11/2012 10:19:36 AM -- Performance logging started (interval: 30000 ms).

5/11/2012 10:19:36 AM -- Backing up databases ...

5/11/2012 2:25:41 PM -- Performance logging has ended.

5/11/2012 2:25:41 PM -- Instance2804.1 (100% processed), Instance2804.2 (100% processed), Instance2804.3 (100% processed), Instance2804.4 (100% processed), Instance2804.5 (100% processed), Instance2804.6 (100% processed), Instance2804.7 (100% processed) and Instance2804.8 (100% processed)

5/11/2012 2:25:41 PM -- C:\Program Files\Exchange Jetstress\DatabaseBackup_2012_5_11_10_19_21.blg has 483 samples.

5/11/2012 2:25:41 PM -- Creating test report ...

Appendix D: Software Recovery Testing

Microsoft Exchange Jetstress 2010
SoftRecovery Test Result Report
Soft-Recovery Statistics - All

Database Instance

Log files replayed

Elapsed seconds

Instance2740.1

507

1418.822492

Instance2740.2

506

1296.2998768

Instance2740.3

509

1713.4602096

Instance2740.4

508

1429.1653102

Instance2740.5

509

1619.8444451

Instance2740.6

501

1647.1600931

Instance2740.7

506

1298.1562801

Instance2740.8

504

1089.708714

Database Configuration

Instance2740.1

Log path: c:\log1

Database: F:\db1\Jetstress001001.edb

Instance2740.2

Log path: c:\log2

Database: F:\db2\Jetstress002001.edb

Instance2740.3

Log path: c:\log3

Database: F:\db3\Jetstress003001.edb

Instance2740.4

Log path: c:\log4

Database: F:\db4\Jetstress004001.edb

Instance2740.5

Log path: c:\log5

Database: F:\db5\Jetstress005001.edb

Instance2740.6

Log path: c:\log6

Database: F:\db6\Jetstress006001.edb

Instance2740.7

Log path: c:\log7

Database: F:\db7\Jetstress007001.edb

Instance2740.8

Log path: c:\log8

Database: f:\db8\Jetstress008001.edb

Transactional I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance2740.1

16.627

0.909

294.811

2.138

41482.607

32768.000

Instance2740.2

14.841

0.877

329.504

2.337

41441.743

32659.497

Instance2740.3

21.041

0.770

214.842

1.774

42088.596

26460.160

Instance2740.4

16.198

0.905

298.760

2.131

41385.913

32768.000

Instance2740.5

18.710

0.921

236.253

1.878

41375.982

31901.122

Instance2740.6

19.247

0.879

237.881

1.817

41265.385

30976.000

Instance2740.7

15.262

0.891

325.591

2.341

41373.858

32659.855

Instance2740.8

13.782

0.877

384.061

2.776

41055.820

32768.000

Transactional I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance2740.1

4.575

0.000

3.207

0.000

232479.930

0.000

Instance2740.2

4.649

0.000

3.504

0.000

231704.650

0.000

Instance2740.3

3.807

0.000

2.662

0.000

187798.541

0.000

Instance2740.4

4.634

0.000

3.195

0.000

232546.742

0.000

Instance2740.5

4.522

0.000

2.818

0.000

223190.830

0.000

Instance2740.6

4.403

0.000

2.726

0.000

219250.801

0.000

Instance2740.7

4.681

0.004

3.511

0.002

231784.251

1.690

Instance2740.8

4.837

0.005

4.164

0.003

232543.793

2.016

Background Database Maintenance I/O Performance

MSExchange Database ==> Instances

Database Maintenance IO Reads/sec

Database Maintenance IO Reads Average Bytes

Instance2740.1

22.936

261191.242

Instance2740.2

24.321

261186.054

Instance2740.3

20.869

261368.039

Instance2740.4

23.994

261357.149

Instance2740.5

22.552

261436.144

Instance2740.6

21.103

261065.191

Instance2740.7

23.087

261077.373

Instance2740.8

24.963

260975.086

Total I/O Performance

MSExchange Database ==> Instances

I/O Database Reads Average Latency (msec)

I/O Database Writes Average Latency (msec)

I/O Database Reads/sec

I/O Database Writes/sec

I/O Database Reads Average Bytes

I/O Database Writes Average Bytes

Instance2740.1

16.627

0.909

317.747

2.138

57341.873

32768.000

Instance2740.2

14.841

0.877

353.825

2.337

56546.327

32659.497

Instance2740.3

21.041

0.770

235.711

1.774

61502.431

26460.160

Instance2740.4

16.198

0.905

322.754

2.131

57738.708

32768.000

Instance2740.5

18.710

0.921

258.805

1.878

60552.027

31901.122

Instance2740.6

19.247

0.879

258.984

1.817

59175.197

30976.000

Instance2740.7

15.262

0.891

348.678

2.341

55920.986

32659.855

Instance2740.8

13.782

0.877

409.024

2.776

54477.404

32768.000

Total I/O Performance continued

MSExchange Database ==> Instances

I/O Log Reads Average Latency (msec)

I/O Log Writes Average Latency (msec)

I/O Log Reads/sec

I/O Log Writes/sec

I/O Log Reads Average Bytes

I/O Log Writes Average Bytes

Instance2740.1

4.575

0.000

3.207

0.000

232479.930

0.000

Instance2740.2

4.649

0.000

3.504

0.000

231704.650

0.000

Instance2740.3

3.807

0.000

2.662

0.000

187798.541

0.000

Instance2740.4

4.634

0.000

3.195

0.000

232546.742

0.000

Instance2740.5

4.522

0.000

2.818

0.000

223190.830

0.000

Instance2740.6

4.403

0.000

2.726

0.000

219250.801

0.000

Instance2740.7

4.681

0.004

3.511

0.002

231784.251

1.690

Instance2740.8

4.837

0.005

4.164

0.003

232543.793

2.016

Host System Performance

Counter

Average

Minimum

Maximum

% Processor Time

8.094

0.000

21.678

Available MBytes

45264.683

45211.000

47040.000

Free System Page Table Entries

33555543.965

33555543.000

33555544.000

Transition Pages RePurposed/sec

0.000

0.000

0.000

Pool Nonpaged Bytes

87252797.925

87240704.000

87322624.000

Pool Paged Bytes

124016175.242

123985920.000

124043264.000

Database Page Fault Stalls/sec

0.000

0.000

0.000

Test Log

5/15/2012 3:41:57 PM -- Jetstress testing begins ...

5/15/2012 3:41:58 PM -- Preparing for testing ...

5/15/2012 3:42:06 PM -- Attaching databases ...

5/15/2012 3:42:06 PM -- Preparations for testing are complete.

5/15/2012 3:42:07 PM -- Starting transaction dispatch ..

5/15/2012 3:42:07 PM -- Database cache settings: (minimum: 256.0 MB, maximum: 2.0 GB)

5/15/2012 3:42:07 PM -- Database flush thresholds: (start: 20.5 MB, stop: 40.9 MB)

5/15/2012 3:42:15 PM -- Database read latency thresholds: (average: 20 msec/read, maximum: 100 msec/read).

5/15/2012 3:42:15 PM -- Log write latency thresholds: (average: 10 msec/write, maximum: 100 msec/write).

5/15/2012 3:42:22 PM -- Operation mix: Sessions 3, Inserts 40%, Deletes 20%, Replaces 5%, Reads 35%, Lazy Commits 70%.

5/15/2012 3:42:22 PM -- Performance logging started (interval: 15000 ms).

5/15/2012 3:42:22 PM -- Generating log files ...

5/15/2012 5:49:58 PM -- c:\log1 (101.4% generated), c:\log2 (101.2% generated), c:\log3 (101.8% generated), c:\log4 (101.6% generated), c:\log5 (101.8% generated), c:\log6 (100.2% generated), c:\log7 (101.2% generated) and c:\log8 (100.8% generated)

5/15/2012 5:49:58 PM -- Performance logging has ended.

5/15/2012 5:49:58 PM -- JetInterop batch transaction stats: 22087, 21848, 21974, 22249, 21877, 22037, 21848 and 22092.

5/15/2012 5:49:59 PM -- Dispatching transactions ends.

5/15/2012 5:49:59 PM -- Shutting down databases ...

5/15/2012 5:50:18 PM -- Instance2740.1 (complete), Instance2740.2 (complete), Instance2740.3 (complete), Instance2740.4 (complete), Instance2740.5 (complete), Instance2740.6 (complete), Instance2740.7 (complete) and Instance2740.8 (complete)

5/15/2012 5:50:18 PM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_15_15_42_15.blg has 493 samples.

5/15/2012 5:50:18 PM -- Creating test report ...

5/15/2012 5:50:22 PM -- Instance2740.1 has 17.6 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.1 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.1 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.2 has 16.9 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.2 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.2 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.3 has 17.6 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.3 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.3 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.4 has 16.9 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.4 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.4 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.5 has 17.4 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.5 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.5 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.6 has 17.5 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.6 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.6 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.7 has 17.2 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.7 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.7 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.8 has 17.2 for I/O Database Reads Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.8 has 1.0 for I/O Log Writes Average Latency.

5/15/2012 5:50:22 PM -- Instance2740.8 has 1.0 for I/O Log Reads Average Latency.

5/15/2012 5:50:22 PM -- Test has 0 Maximum Database Page Fault Stalls/sec.

5/15/2012 5:50:22 PM -- The test has 0 Database Page Fault Stalls/sec samples higher than 0.

5/15/2012 5:50:22 PM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_15_15_42_15.xml has 492 samples queried.

5/15/2012 5:50:23 PM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_15_15_42_15.html was saved.

5/15/2012 5:50:29 PM -- Performance logging started (interval: 4000 ms).

5/15/2012 5:50:29 PM -- Recovering databases ...

5/15/2012 6:19:02 PM -- Performance logging has ended.

5/15/2012 6:19:02 PM -- Instance2740.1 (1418.822492), Instance2740.2 (1296.2998768), Instance2740.3 (1713.4602096), Instance2740.4 (1429.1653102), Instance2740.5 (1619.8444451), Instance2740.6 (1647.1600931), Instance2740.7 (1298.1562801) and Instance2740.8 (1089.708714)

5/15/2012 6:19:03 PM -- C:\Program Files\Exchange Jetstress\SoftRecovery_2012_5_15_17_50_23.blg has 401 samples.

5/15/2012 6:19:03 PM -- Creating test report ...

Appendix E: Database Checksum

Microsoft Exchange Jetstress 2010
Test Result Report
Checksum Statistics - All

Database

Seen pages

Bad pages

Correctable pages

Wrong page-number pages

File length / seconds taken

F:\db1\Jetstress001001.edb

13710370

0

0

0

428449 MB/709 sec

F:\db2\Jetstress002001.edb

13711394

0

0

0

428481 MB/583 sec

F:\db3\Jetstress003001.edb

13710626

0

0

0

428457 MB/569 sec

F:\db4\Jetstress004001.edb

13709858

0

0

0

428433 MB/584 sec

F:\db5\Jetstress005001.edb

13711138

0

0

0

428473 MB/597 sec

F:\db6\Jetstress006001.edb

13709858

0

0

0

428433 MB/611 sec

F:\db7\Jetstress007001.edb

13710114

0

0

0

428441 MB/631 sec

f:\db8\Jetstress008001.edb

13710114

0

0

0

428441 MB/990 sec

(Sum)

109683472

0

0

0

3427608 MB/4288 sec

Disk Subsystem Performance (of checksum)

LogicalDisk

Avg. Disk sec/Read

Avg. Disk sec/Write

Disk Reads/sec

Disk Writes/sec

Avg. Disk Bytes/Read

F:

0.013

0.000

12767.216

0.000

65536.000

f:

0.011

0.000

17091.822

0.000

65536.000

Memory System Performance (of checksum)

Counter

Average

Minimum

Maximum

% Processor Time

3.392

1.838

5.478

Available MBytes

47205.043

47175.000

47211.000

Free System Page Table Entries

33555543.454

33555543.000

33555545.000

Transition Pages RePurposed/sec

0.000

0.000

0.000

Pool Nonpaged Bytes

87869890.270

87834624.000

87949312.000

Pool Paged Bytes

119515760.567

119435264.000

119803904.000

Test Log

5/29/2012 10:12:53 AM -- Jetstress testing begins ...

5/29/2012 10:12:53 AM -- Preparing for testing ...

5/29/2012 10:13:02 AM -- Attaching databases ...

5/29/2012 10:13:02 AM -- Preparations for testing are complete.

5/29/2012 10:13:03 AM -- Starting transaction dispatch ..

5/29/2012 10:13:03 AM -- Database cache settings: (minimum: 256.0 MB, maximum: 2.0 GB)

5/29/2012 10:13:03 AM -- Database flush thresholds: (start: 20.5 MB, stop: 40.9 MB)

5/29/2012 10:13:11 AM -- Database read latency thresholds: (average: 20 msec/read, maximum: 100 msec/read).

5/29/2012 10:13:11 AM -- Log write latency thresholds: (average: 10 msec/write, maximum: 100 msec/write).

5/29/2012 10:13:21 AM -- Operation mix: Sessions 3, Inserts 40%, Deletes 20%, Replaces 5%, Reads 35%, Lazy Commits 70%.

5/29/2012 10:13:21 AM -- Performance logging started (interval: 15000 ms).

5/29/2012 10:13:21 AM -- Attaining prerequisites:

5/29/2012 10:17:24 AM -- \MSExchange Database(JetstressWin)\Database Cache Size, Last: 1941848000.0 (lower bound: 1932735000.0, upper bound: none)

5/30/2012 10:17:24 AM -- Performance logging has ended.

5/30/2012 10:17:24 AM -- JetInterop batch transaction stats: 222502, 223361, 223068, 222069, 223510, 222601, 222702 and 222837.

5/30/2012 10:17:24 AM -- Dispatching transactions ends.

5/30/2012 10:17:24 AM -- Shutting down databases ...

5/30/2012 10:17:44 AM -- Instance2536.1 (complete), Instance2536.2 (complete), Instance2536.3 (complete), Instance2536.4 (complete), Instance2536.5 (complete), Instance2536.6 (complete), Instance2536.7 (complete) and Instance2536.8 (complete)

5/30/2012 10:17:44 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.blg has 5614 samples.

5/30/2012 10:17:44 AM -- Creating test report ...

5/30/2012 10:18:59 AM -- Instance2536.1 has 15.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.1 has 1.1 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.1 has 1.1 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 14.8 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 1.1 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.2 has 1.1 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 15.4 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.3 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 16.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.4 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 17.2 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.5 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 16.4 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.6 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 15.3 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.7 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 14.9 for I/O Database Reads Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 1.2 for I/O Log Writes Average Latency.

5/30/2012 10:18:59 AM -- Instance2536.8 has 1.2 for I/O Log Reads Average Latency.

5/30/2012 10:18:59 AM -- Test has 0 Maximum Database Page Fault Stalls/sec.

5/30/2012 10:18:59 AM -- The test has 0 Database Page Fault Stalls/sec samples higher than 0.

5/30/2012 10:18:59 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.xml has 5598 samples queried.

5/30/2012 10:18:59 AM -- C:\Program Files\Exchange Jetstress\Performance_2012_5_29_10_13_11.html was saved.

5/30/2012 10:19:00 AM -- Performance logging started (interval: 30000 ms).

5/30/2012 10:19:00 AM -- Verifying database checksums ...

5/30/2012 11:30:29 AM -- F: (100% processed) and f: (100% processed)

5/30/2012 11:30:29 AM -- Performance logging has ended.

5/30/2012 11:30:29 AM -- C:\Program Files\Exchange Jetstress\DBChecksum_2012_5_30_10_18_59.blg has 141 samples.

Text Box: Printed in USA	C11-722070-00	11/12