HP Hewlett Packard Personal Computer BL465C G7 User Manual

Performance of HP ProLiant BL465c G7  
with AMD Opteron 6100 Series  
processors in 32- and 64-bit HP SBC  
environments  
Technical white paper  
Table of contents  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Introduction  
Figure 1 shows the HP ProLiant BL465c G7 server blade.  
Figure 1. The HP ProLiant BL465c G7 server blade, illustrating a unique drawer design that allows you to service hard drives  
independently  
G7 technologies  
Seventh-generation AMD Opteron-based HP ProLiant servers such as the BL465c G7 server blade  
feature the following key technologies:  
AMD Opteron 6100 Series processors  
DDR3 memory  
Embedded dual-port FlexFabric 10 Gb Ethernet converged network adapter4  
PCI Express 2.0  
Smart Array controller with 1 GB Flash Backed Write Cache (FBWC)  
FBWC  
The embedded Smart Array P410i RAID controller5 featured in the HP  
ProLiant BL465c G7 server blade provides 1 GB FBWC as standard.  
FBWC, which does not require a battery, uses flash memory.  
Data retention is indefinite; by comparison, retention with Battery Backed  
Write Cache (BBWC) is approximately two days.  
4 Currently, FlexFabric is only available on HP ProLiant c-Class server blades.  
5 FBWC may be optional on some models. For more information on FBWC, refer to  
3
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
HP Integrated Lights-Out 3 (iLO 3)  
Serial Attached SCSI (SAS)  
Thermal logic  
The following sections of this paper describe testing performed by HP to characterize the  
performance and scalability of an HP ProLiant BL465c G7 server blade in 32- and 64-bit HP SBC  
environments.  
Test methodology  
HP continues to upgrade existing HP ProLiant servers and introduce new servers to meet particular  
business needs. To help you select the appropriate server for your particular HP SBC environment, HP  
publishes this and other performance characterizations so that you can compare individual server  
performance and scalability.  
This section describes how HP determined the optimal number of users supported by a 2P HP ProLiant  
BL465c G7 server blade featuring a range of AMD Opteron Model 6100 processors henceforth  
referred to as the HP ProLiant BL465c G7 server blade in 64- and 32-bit test harnesses.  
Important:  
As with any laboratory testing, the performance metrics quoted in this  
paper are idealized. In a production environment, these metrics may be  
impacted by a variety of factors.  
HP recommends proof-of-concept testing in a non-production environment  
using the actual target application as a matter of best practice for all  
application deployments. Testing the actual target application in a  
test/staging environment identical to, but isolated from, the production  
environment is the most effective way to characterize system behavior.  
Note:  
A 64-bit HP SBC environment eliminates the kernel memory constraints that  
can limit server scalability in a 32-bit HP SBC environment. For more  
x64 Editions.  
This section provides more information on test tools, user profile and test scenarios.  
Test tools  
To facilitate the placement and management of simulated loads on an HP SBC server, HP used  
Terminal Services Scalability Planning Tools (TSScaling), a suite of tools developed by Microsoft® to  
help organizations with Microsoft Windows® Server 2003 Terminal Server capacity planning.  
Table 1 describes these tools.  
4
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Table 1. Components of TSScaling  
Component  
Description  
Automation tools  
Robosrv.exe  
Drives the server-side of the load simulation  
Helps drive the client-side of the load simulation  
Robocli.exe  
Qidle.exe  
Test tools  
Determines if any scripts have failed and require  
operator intervention  
Tbscript.exe  
A script interpreter that helps drive the client-side load  
simulation  
Help files  
TBScript.doc  
Terminal Server bench scripting documentation  
A scalability test environment set-up guide  
A testing guide  
TSScalingSetup.doc  
TSScalingTesting.doc  
More information  
Roboserver (Robosrv.exe) and Roboclient (Robocli.exe): Terminal Server capacity planning  
User profile  
To simulate a typical workload in the HP SBC test environment, HP used a script based on the  
following Heavy User profile:  
Heavy Users (also known as Structured Task Workers) tend to open multiple applications  
simultaneously and remain active for long periods. Heavy Users often leave applications  
open when not in use.  
Table 2 outlines the activities performed by Heavy Users, which utilized Microsoft Office 2003  
products.  
Table 2. Activities incorporated into the test script  
Activity  
Description  
Access  
Open a database, apply a filter, search through records, add records, and delete  
records.  
Excel  
Open, print and save a large spreadsheet.  
InfoPath  
Outlook  
Enter data6 into a form; save the form over an existing form.  
First pass: Email a short message.  
Second pass: Email a reply with an attachment.  
Outlook_2  
PowerPoint  
Create a long reply.  
Create a new presentation, insert clipart, and apply animation. View the  
presentation after each slide is created.  
PowerPoint_2  
Word  
Open and view a large presentation with heavy animation and many colors and  
gradients.  
Create, save, print, and email a document.  
6 Data entry for Office InfoPath 2003 requires significant processor resources.  
5
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Test scenario  
To characterize its scalability, HP tested the HP ProLiant BL465c G7 server blade in 64- and 32-bit HP  
SBC environments when running a workload based on the User profile described above.  
Note:  
As recommended by HP for most applications, the FBWC was configured  
with 100% write for all testing. Use the BIOS to adjust the read/write ratio.  
Testing was initiated by running the particular workload with a group of 15 simulated users; start  
times were staggered to eliminate authentication overhead. After these sessions finished, HP added  
15 more users, then repeated the testing. Further users were added until the optimal number (as  
described below) was reached.  
Performance and scalability metrics  
While the Heavy User workload was running, HP monitored a range of Windows Performance  
Monitor (Perfmon) counters in order to characterize the performance and scalability of the tested  
server. HP also used canary scripts featuring Office 2003-based activities to establish the number of  
users that could be supported before response times became unacceptable.  
HP typically uses the Perfmon % Processor Time counter to establish the optimal number of users  
supported by an HP SBC server by definition, the number of users active when processor utilization  
reaches 80%. At this time, a limited number of additional users or services can be supported;  
however, user response times may become unacceptable.  
In a 32-bit HP SBC environment, System Page Table Entries (PTEs) may become exhausted7 before  
processor utilization reaches 80%.  
To validate metrics obtained from Perfmon, HP uses canary scripts to characterize response times for  
a range of discrete activities, such as the time taken to invoke an application or for a modal box to  
appear. By monitoring response times a very practical metric as more and more users log on, HP  
has been able to demonstrate that these times are acceptable when the optimal number of users (as  
determined using Perfmon counter values) is active.  
Note:  
When running canary scripts, HP considers user response times to become  
unacceptable when they increase markedly over a baseline measurement.  
7 Due to inherent limitations in the x86 platform; for more information, refer to Appendix A Using Microsoft Windows Server 2003 x64  
6
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Test topology  
Figure 2 illustrates the HP SBC test environment.  
Figure 2. The tested environment (64-bit implementation shown)  
Note:  
Test environments such as that shown in Figure 2 are available to customers  
at HP Solution Centers to help solve a wide variety of business problems.  
7
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Configurations  
This section outlines the configurations of servers and clients used in the test environment.  
Table 3 presents information on system configurations; Table 4 provides a system summary.  
Table 3. System configurations  
Server  
Configuration  
HP SBC server  
2P HP ProLiant BL465c G7 server blade with:  
AMD Opteron processor Model 6174 (12-core/2.2 GHz), Model 6164 HE (12-  
core/1.7 GHz), Model 6136 (8-core/2.4 GHz), or Model 6128 HE (8-core/2.0  
GHz)  
12 MB shared L3 cache  
64 GB RAM (8 GB for Model 6136, 32-bit testing)  
Smart Array P410i controller with RAID 0  
Two 146 GB 10,000 rpm SAS hard drives  
96 GB page file on system partition (12 GB for Model 6136, 32-bit testing)  
1 GB FBWC (100% write)  
HP NC551i Dual Port FlexFabric 10 Gb Ethernet Converged Network Adapter  
64-bit: Windows Server 2003 R2 Enterprise x64 Edition with Service Pack 2  
32-bit: Windows Server 2003 Enterprise x86 Edition with Service Pack 1  
Terminal Services enabled  
Office 2003  
Exchange Server/  
Internet Information Services  
2P HP ProLiant DL360 G5 server with:  
Dual-core Intel® Xeon® processor (3.2 GHz)  
2 x 2 MB L2 cache  
2 GB RAM  
Four 72 GB 15,000 rpm SAS hard drives  
Integrated Smart Array P400i controller with RAID 5  
NC373i Multifunction Gigabit Server Adapter  
Windows Server 2003 Enterprise Edition  
Microsoft Exchange Server 2003  
Microsoft Internet Information Services (IIS) 6.0  
Continued  
8
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Table 3. System configurations (continued)  
Server  
Configuration  
Domain controller  
2P HP ProLiant DL360 G5 server with:  
Dual-core Intel Xeon processor (3.2 GHz)  
2 x 2 MB L2 cache  
2 GB RAM  
Four 72 GB 15,000 rpm SAS hard drives  
Integrated Smart Array P400i controller with RAID 5  
NC373i Multifunction Gigabit Server Adapter  
Windows Server 2003 Enterprise Edition  
Client  
Variety of Intel Pentium®-based Compaq Evo workstations (600 MHz 2.533 GHz),  
each with:  
At least 256 MB of memory  
1024 x 768/16-bit color depth  
100 Mbps NIC  
Windows 2000 Professional or Windows XP  
Table 4. System summary for the bare-metal HP ProLiant BL465c G7 server blade (listings shown in bold are unique to the 32-  
bit test environment)  
Component  
Description  
Operating system  
Microsoft Windows Server 2003, Enterprise x64 Edition/  
Microsoft Windows Server 2003, Enterprise Edition  
Version  
5.2.3790 Service Pack 2, Build 3790/  
5.2.3790 Service Pack 1, Build 3790  
Other OS description  
System model  
R2/Not applicable  
ProLiant BL465 G7  
System type  
x64-based PC/x86-based PC  
Processor each of 24 cores  
(Model 6174 or 6164 HE)  
or 16 cores (Model 6136 or  
6128 HE)  
AMD64/x86 Family 16 Model 9 Stepping 1 Authentic AMD  
Model 6174: ~2200 MHz  
Model 6164 HE: ~1700 MHz  
Model 6136: ~2400 MHz  
Model 6128 HE: ~2000 MHz  
BIOS version/date  
SMBIOS version  
Windows directory  
System directory  
HP A19, 1/22/2010  
2.6  
C:\WINDOWS  
C:\WINDOWS\system32  
Continued  
9
Download from Www.Somanuals.com. All Manuals Search And Download.  
Table 4. System summary for the bare-metal HP ProLiant BL465c G7 server blade (continued)  
Component  
Description  
Boot device  
\Device\HarddiskVolume1  
United States  
Locale  
Hardware abstraction layer  
5.2.3790.3959 (srv03_sp2_rtm.070216-1710)/  
5.2.3790.1830 (srv03_sp1_rtm.050324-1447)  
User name  
Not available  
Total physical memory  
Model 6174: 65,531.39 MB/65,531.39 MB  
Model 6164 HE: 65,531.39 MB  
Model 6136: 65,531.39 MB /8,187.39 MB  
Model 6128 HE: 65,531.39 MB  
Available physical memory  
Total virtual memory  
Available virtual memory  
Page file space  
Model 6174: 61.96 GB/3.25 GB  
Model 6164 HE: 62.06 GB  
Model 6136: 61.95 GB/3.56 GB  
Model 6128 HE: 62.03 GB  
Model 6174: 158.42 GB/3.25 GB  
Model 6164 HE: 158.42 GB  
Model 6136: 158.42 GB/3.80 GB  
Model 6128 HE:158.42 GB  
Model 6174: 157.98 GB/3.25 GB  
Model 6164 HE: 158.06 GB  
Model 6136: 157.97 GB/3.56 GB  
Model 6128 HE: 158.04 GB  
Model 6174: 95.99 GB /95.99 GB  
Model 6164 HE: 95.99 GB  
Model 6136: 95.99 GB /11.99 GB  
Model 6128 HE: 95.99 GB  
Page file  
C:\pagefile.sys  
Performance test results  
This section outlines the test results used by HP to characterize the optimal performance and scalability  
of the HP ProLiant BL465c G7 server blade. The following metrics were used:  
Perfmon values Shows select Perfmon counter values for the Heavy User workload  
Canary times Shows user response times for a sample canary script; used to validate optimal  
scalability levels  
Note:  
As with any laboratory benchmark, the performance metrics quoted in this  
performance brief are idealized. In a production environment, these metrics  
may be impacted by a variety of factors; for more information, refer to  
HP determined that there were no disk or network bottlenecks in the test  
environment.  
10  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
The following scenarios were tested:  
Server blade configured with the AMD Opteron processor Model 6174, 6164 HE, 6136, and  
6128 HE  
HP SBC test environments  
64-bit: 6174, 6164 HE, 6136, 6128 HE  
32-bit: 6174, 6136  
Configured with processor Model 6174  
An HP ProLiant BL465c G7 server blade featuring the AMD Opteron processor Model 6174 was  
tested in 64- and 32-bit HP SBC test environments.  
64-bit test environment  
Perfmon values and canary times are presented.  
Perfmon values  
HP ran a performance test using a workload based on the Heavy User profile. Figure 3 shows select  
Perfmon values during this test run; Figure 4 presents normalized results.  
Figure 3. % Processor Time values showing that 487 Heavy Users were supported when processor utilization reached 80%  
Figure 3 shows the optimal number of Heavy Users supported by the bare-metal HP ProLiant BL465c  
G7 server blade to be 487.  
HP noted that, as the test run progressed, processor queue length started to increase steadily and  
never emptied. With approximately 500 users logged on (that is, more than the optimal number),  
there was a marked increase in queue length.  
11  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Figure 4 shows normalized Perfmon values.  
Figure 4. There were no stopped sessions when 487 Heavy Users were logged on  
With 487 users active, there were no stopped sessions.  
The processor became saturated with approximately 500 users logged on, which coincided with the  
marked increase in processor queue length shown in Figure 3.  
The maximum number of users able to log on was 739, though approximately 2% of sessions had  
stopped.  
12  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Canary times  
Figure 5 shows sample results for the tested server when running a typical canary script. Individual  
user response times are shown in blue, with a yellow line depicting average response times.  
HP analyzed this figure to determine when response times began to increase markedly and  
consistently over a baseline level, indicating that user response times had become unacceptable.  
Figure 5. Canary time values show that user response times started to become unacceptable when 514 Heavy Users were  
active  
Figure 5 indicates that response times were acceptable when 487 Heavy Users the optimal number  
were active.  
However, more users were able to log on up to a maximum of 739, limited by lack of CPU resources.  
32-bit test environment  
HP ran a performance test using a workload based on the Heavy User profile. Figure 6 shows select  
Perfmon values during this test run.  
13  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Figure 6. % Processor Time values showing that 141 Heavy Users were supported when system PTEs became exhausted  
The optimal number of Heavy Users supported by the HP ProLiant BL465c G7 server blade was 141.  
Configured with processor Model 6164 HE  
An HP ProLiant BL465c G7 server blade featuring the AMD Opteron processor Model 6164 HE was  
tested in the 64-bit HP SBC test environment.  
64-bit test environment  
Perfmon values and canary times are presented.  
Perfmon values  
HP ran a performance test using a workload based on the Heavy User profile. Figure 7 shows select  
Perfmon values during this test run; Figure 8 presents normalized results.  
14  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Figure 7. % Processor Time values showing that 447 Heavy Users had been able to log on when processor utilization  
reached 80%  
Thus, as many as 447 Heavy Users were supported when processor utilization reached 80%.  
Figure 8 shows normalized Perfmon values.  
15  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Figure 8. Although 447 Heavy Users were able to log on, two sessions had stopped  
With 447 Heavy Users the number of users logged on when processor utilization reached 80% –  
there were two stopped sessions. Thus the optimal number of users is (447 2); that is, 445.  
16  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Canary times  
Figure 9 shows sample results for the tested server when running a typical canary script. Individual  
user response times are shown in blue, with a yellow line depicting average response times.  
HP analyzed this figure to determine when response times began to increase markedly and  
consistently over a baseline level, indicating that user response times had become unacceptable.  
Figure 9. Canary time values show that user response times started to become unacceptable when 477 Heavy Users were  
active  
Response times were acceptable when 445 Heavy Users the optimal number were active.  
Configured with processor Model 6136  
An HP ProLiant BL465c G7 server blade featuring the AMD Opteron processor Model 6136 was  
tested in 64- and 32-bit HP SBC test environments.  
64-bit test environment  
Perfmon values and canary times are presented.  
Perfmon values  
HP ran a performance test using a workload based on the Heavy User profile. Figure 10 shows select  
Perfmon values during this test run.  
17  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Figure 10. % Processor Time values showing that 411 Heavy Users were supported when processor utilization reached 80%  
The optimal number of Heavy Users supported by the bare-metal HP ProLiant BL465c G7 server blade  
was 411.  
Canary times  
Figure 11 shows sample results for the tested server when running a typical canary script. Individual  
user response times are shown in blue, with a yellow line depicting average response times.  
HP analyzed this figure to determine when response times began to increase markedly and  
consistently over a baseline level, indicating that user response times had become unacceptable.  
18  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Figure 11. Canary time values show that user response times started to become unacceptable when 475 Heavy Users were  
active  
Response times were acceptable when 411 Heavy Users the optimal number were active.  
32-bit test environment  
HP ran a performance test using a workload based on the Heavy User profile. Figure 12 shows select  
Perfmon values during this test run.  
19  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Figure 12. % Processor Time values showing that 161 Heavy Users were supported when disk utilization began to increase  
exponentially  
Disk utilization increased exponentially during this test run due to lack of system PTEs.  
The optimal number of users supported by this HP ProLiant BL465c G7 server blade was 161.  
Configured with processor Model 6128 HE  
An HP ProLiant BL465c G7 server blade featuring the AMD Opteron processor Model 6128 HE was  
tested in a 64-bit HP SBC test environment.  
64-bit test environment  
Perfmon values and canary times are presented.  
Perfmon values  
HP ran a performance test using a workload based on the Heavy User profile. Figure 13 shows select  
Perfmon values during this test run.  
20  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Figure 13. % Processor Time values showing that 380 Heavy Users were supported when processor utilization reached 80%  
The optimal number of Heavy Users supported by the bare-metal HP ProLiant BL465c G7 server blade  
was 380.  
Canary times  
Figure 14 shows sample results for the tested server when running a typical canary script. Individual  
user response times are shown in blue, with a yellow line depicting average response times.  
HP analyzed this figure to determine when response times began to increase markedly and  
consistently over a baseline level, indicating that user response times had become unacceptable.  
21  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Figure 14. Canary time values show that user response times started to become unacceptable when 406 Heavy Users were  
active  
Response times were acceptable when 380 Heavy Users the optimal number were active.  
Test analysis summary  
Figure 15 summarizes the test results.  
Figure 15. Optimal numbers of Heavy Users supported by the HP ProLiant BL465c G7 server blade in an HP SBC environment  
22  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Important:  
As with any laboratory benchmark, the performance metrics quoted in this  
performance brief are idealized. In a production environment, these metrics  
may be impacted by a variety of factors; for more information, refer to  
HP determined that there were no disk or network bottlenecks in the test  
environment.  
Comparing scalability in the 32-bit environment  
Scalability in the 32-bit HP SBC test environment was impacted by the particular processor model  
being used, as shown in Figure 16.  
Figure 16. More users were supported when the tested server was equipped with 8-core processors  
Because the amount of memory consumed by chip-set drivers increases with core density, fewer free  
System PTEs were available to the Model 6174 (12-core) processor-powered server at the start of the  
test, which translated to support for 38% fewer Heavy Users.  
23  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Recommendations  
This section provides general recommendations for improving the performance of HP SBC servers.  
x64 platforms  
Since x64 platforms allow you to better utilize memory and multi-core processors, the bottleneck you  
are most likely to encounter8 would be associated with the disk subsystem. While a detailed analysis  
of disk I/O performance is beyond the scope of this white paper, the following observations are  
offered to help you improve disk performance:  
Utilize write cache (see Using write cache)  
Since internal storage is often insufficient to support a large number of users in an HP SBC  
environment, consider deploying additional RAID arrays and/or SAN support. Note also that when  
a SCSI RAID array is used to host user profiles and page files, the number of spindles deployed has  
a significant impact on the response times associated with file access.  
When the pressure on the disk I/O subsystem is high, one option for improving disk access times is  
to add RAM to lower the pressure on memory.  
As the number of sessions increases, disk activity and the pressure on the disk I/O subsystem also  
increase. If file I/O activity is high, the probability that requests will find the desired data in  
memory decreases, thus negatively affecting file access times.  
Avoiding disk I/O bottlenecks  
To help you avoid disk I/O bottlenecks, Microsoft recommends using the Windows performance  
monitoring tool, Perfmon, to check the following metrics9:  
%Idle time Idle times for logical and physical drives should average at least 50%  
Average Disk Seconds/Read and Average Disk Seconds/Write The average time taken to  
complete a read or write should average less than 25 milliseconds, with peaks less than 50  
milliseconds  
If the above conditions specified by Microsoft cannot be met, a disk I/O bottleneck is likely.  
Note:  
In the event of an I/O bottleneck, you should tune the disk subsystem,  
decrease the number of users or applications, or add memory to the server.  
Using write cache  
HP Smart Array controllers include an allocation of memory that can be utilized to temporarily cache  
data being written to or read from disk. Since access to this memory is significantly faster than disk  
access, cache can enhance overall server performance, particularly during login operations.  
Write cache is of particular interest in an HP SBC environment. After buffering all the data associated  
with a particular write command, the Smart Array controller indicates to the HP SBC server that the  
data transfer to the disk is complete even through the data is still being written to disk. This frees up  
the server’s processor to perform other tasks and accelerates data throughput.  
8 For further information, refer to the HP white paper, “Scalability and performance of HP ProLiant servers on 64-bit Microsoft  
Windows Server 2003 in an HP SBC environment.”  
9 For further information, visit the Microsoft website.  
24  
Download from Www.Somanuals.com. All Manuals Search And Download.  
       
Note:  
HP has not yet characterized FBWC performance in the HP SBC  
environment. However, testing performed using BBWC demonstrates that,  
typically, improvements from write cache are most significant when the HP  
SBC server is performing log-intensive operations and/or when significant  
page file write operations are necessary, such as during user logins.  
Performance gains ranged from 50% to 250%10; actual results would vary  
depending on the application(s) involved and your particular HP SBC  
environment.  
FBWC is enabled by default. Use the BIOS to set the read/write ratio11.  
Virtualization  
To take best advantage of the benefits delivered by virtualization, you need to understand your HP  
SBC environment. In addition, be aware that virtual machine (VM) performance may vary depending  
on the application, the guest operating system, and other factors; you should test the VM prior to  
implementation in a production environment. Also be aware that some applications are good  
candidates for virtualization, others less so: for example, underutilized HP SBC servers, servers  
running infrastructure services, and Citrix XenApp data store servers may be good candidates for  
virtualization; HP SBC servers running resource-intensive applications and highly-utilized infrastructure  
servers may not be such good candidates.  
When you are planning a virtualized implementation, ensure you have selected a server with the CPU  
capacity and number of cores you need. Make sure you meet the resource requirements for  
virtualization overhead, guest operating systems, and applications. Would a storage array network  
(SAN) be a better choice than internal storage? Are there enough network interface cards (NICs)?  
To correctly size the HP SBC servers you intend to virtualize, you must understand the associated  
applications and the numbers of users and user profiles to be supported. You should balance the  
number of VMs deployed on a particular server with the number of vCPUs allocated to each VM; you  
should also allocate enough RAM to eliminate memory and I/O bottlenecks. Prior to deployment, HP  
recommends testing your VMs in a production test environment with live users.  
In addition to suitable sizing, optimal VM performance requires Citrix XenServer and guest operating  
systems to be appropriately configured. For example, consider disabling the screen saver associated  
with the XenServer VM controller window.  
To avoid spikes in processor utilization, ensure your VMs are online before applying the workload.  
Do not simultaneously add large numbers of users; if possible, balance the workload across your  
VMs.  
A broad range of tools is available to help you manage a virtualized HP SBC environment, including  
Citrix Essentials for XenServer Solution - HP Edition and Citrix XenCenter.  
VMs are flexible, allowing you to readily implement the level of availability you need. Moreover, you  
can enhance availability by utilizing a storage array network (SAN) created from HP StorageWorks  
product offerings. Capabilities may include:  
Multiple paths for redundancy  
Automatic path failover  
High-availability cluster support  
10 For further information, visit the HP website.  
11 For most applications, HP recommends 100% write.  
25  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Note:  
For more information on best practices associated with virtualization, refer  
XenServer.”  
Conclusions  
When planning an x64 HP SBC environment, you should select servers equipped with multi-core  
processors to help maximize scalability. If your budget allows, consider the fastest processors, the  
most cores, and largest cache; if your objective is to reduce overall power consumption and you are  
prepared to accept a small performance penalty, use low-power processors.  
Memory is an important factor in the x64 environment: while an x64 platform can utilize more RAM,  
it also has a higher minimum RAM requirement than an x86 platform. Since a system that is not  
memory-starved is less likely to experience disk I/O bottlenecks, HP recommends adding as much  
RAM as your budget permits.  
For optimal performance, configure FBWC for 100% write. Consider using 15,000 rpm SAS drives.  
Virtualized server  
Do not oversubscribe vCPUs after a certain point, additional vCPUs can degrade performance.  
Determine the optimal server configuration for your particular workload: How many VMs? How many  
vCPUs? How much memory? Significant testing may be required.  
Consider virtualizing later x86 platforms to take advantage of improvements made to processors,  
memory, and I/O. You may experience dramatic scalability enhancements.  
26  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Appendix A Using Microsoft Windows Server 2003 x64  
Editions  
Microsoft offers operating systems that support both 64- and 32-bit applications, as well as existing  
32-bit deployment and management tools all on the same platform. These operating systems  
provide an evolutionary path to 64-bit technology, allowing 64- and 32-bit applications to run side-  
by-side during the gradual migration to 64-bit computing.  
64-bit editions of Windows Server 2003 running on today’s multi-core AMD Opteron or Intel Xeon  
processors can improve the performance of HP SBC servers by processing more data per clock cycle,  
addressing more memory, and running some numerical calculations faster. Large data sets can be  
loaded entirely into memory, reducing the need for slower disk access; complex calculations that take  
hours to complete on a 32-bit system can be performed in minutes; and workloads that once required  
a large server farm can be performed by a single server.  
In addition, 64-bit platforms also remove many of the limitations that have previously inhibited  
scalability in an HP SBC environment.  
Historical scalability limitations  
32-bit Windows operating systems can directly address 4 GB of memory, 2 GB of which is reserved  
for the operating system kernel and 2 GB for applications. Since kernel memory is shared by all  
applications, the relatively small size of this space can be particularly problematic in an HP SBC  
environment where a server may be responsible for hundreds of users and thousands of processes. In  
this scenario, kernel memory can become constrained, making user response times unacceptably long  
and effectively limiting the ability of the server to scale up.  
Historically, HP SBC environments have been implemented using 1P or 2P servers. Larger, more  
powerful servers have typically not been deployed for two main reasons:  
Kernel memory issues have limited the performance of more powerful servers; either a disk I/O  
bottleneck occurs or kernel memory is consumed before processor resources can be fully utilized  
Scalability in a 32-bit symmetric multi-processing (SMP) system is inherently non-linear above 2P  
With these 1P and 2P server farms, opportunities to scale up are limited. As a result, customers are  
forced to scale out, which can create new problems such as deployment and management  
complexity, high power and cooling requirements, under-utilized resources, and minimal opportunities  
for server consolidation.  
The 64-bit platform shatters the earlier 4 GB limitation for example, Windows Server 2003 R2  
Datacenter x64 Edition with Service Pack (SP) 1 supports up to 2 TB of RAM effectively removing  
kernel memory limitations and eliminating disk I/O bottlenecks. By deploying a Windows Server  
x64 Edition operating system, customers can fully utilize the resources of their existing HP SBC  
servers and take full advantage of new, more powerful systems whether they are running 32- or  
64-bit applications.  
More information  
For more information on the impact of 64-bit Windows Server 2003 x64 Editions in an HP SBC  
environment, refer to the HP white paper, “Scalability and performance of HP ProLiant servers on 64-  
bit Microsoft Windows Server 2003 in an HP SBC environment.”  
To learn about 64-bit computing in an HP SBC environment, refer to the HP white paper,  
“Fundamentals of 64-bit computing in an HP SBC environment.”  
27  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Appendix B SBC solution sizing  
As with any laboratory benchmark, the performance metrics quoted in this performance brief are  
idealized. In a production environment, these metrics may be impacted by a variety of factors,  
including the following:  
Overhead  
Agents and services (virus scanning, backup and restore, provisioning, security, management and  
more) automatically consume overhead. Rogue applications can consume additional overhead.  
The system architect may wish to provide a 25% 30% buffer to accommodate this overhead.  
Future growth  
To accommodate future growth, the system architect may wish to provide an additional buffer.  
Alternatively, servers can be added as needed, taking advantage of the server farm’s inherent  
ability to scale out.  
User profiles  
The particular application in use directly impacts the number of users supported by a particular  
server. Further, user behavior can also impact scalability:  
Increased typing rates correspond to fewer users.  
Opening and closing applications (rather than switching between them) or moving quickly  
between tasks can place a heavier load on the server.  
Sizing for this performance test was based on the Heavy User profile described in User profile. If  
this profile does not match your needs, more profiles are available using the online sizer tool  
(described below); alternatively, the system architect can consult HP Services for more  
information.  
Background grammar checking  
Background grammar checking can significantly impact scalability, reducing the number of users  
supported by as much as 50%. HP disabled background grammar checking for the testing  
described in this performance brief.  
Online sizer tool  
To minimize risk, HP offers automated, downloadable tools that can help you size an HP SBC  
Services) is available for enterprise and small and medium business (SMB) environments.  
The algorithms and methodology used by this sizer are based on the results of customer surveys and  
thorough testing.  
Figure B-1 shows the home page for the sizer.  
28  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Figure B-1. Home page for the HP Citrix XenApp and Microsoft Terminal Services Sizing Tool  
Based on information provided by the customer, the sizer can provide a quick, consistent mechanism  
for identifying the “best-fit” server for a particular HP SBC environment and generate a Bill of  
Materials (BOM) for that server.  
29  
Download from Www.Somanuals.com. All Manuals Search And Download.  
For more information  
HP ProLiant BL465c G7 server blade  
HP ActiveAnswers solution area for HP SBC, http://www.hp.com/solutions/activeanswers/hpsbc  
including Citrix XenApp and Microsoft  
Terminal Services  
HP ProLiant Sizer for Citrix XenApp and  
Microsoft Windows Server 2003 Terminal  
Services  
HP Services  
HP Solution Centers  
Citrix XenApp  
Citrix XenServer  
Citrix Essentials for XenServer Solution - HP  
Edition  
To help us improve our documents, please provide feedback at  
© Copyright 2010 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.  
Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation. Intel, Xeon, and Pentium are  
trademarks of Intel Corporation in the U.S. and other countries. AMD Opteron is a trademark of Advanced  
Micro Devices, Inc  
4AA1-7363ENW, Created June 2010  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 

Greenheck Fan Ventilation Hood OFSD 212 User Manual
Grizzly Sander H0592 User Manual
Grundig Hot Beverage Maker WK 4062 User Manual
Hach Network Router DR 2800 User Manual
Hamilton Beach Coffeemaker 47535C User Manual
Harman Kardon Turntable CDR 26 User Manual
Hasbro Robotics 87776 User Manual
Hotpoint Dishwasher BFI620 User Manual
Husqvarna Chainsaw 1151375 95 User Manual
Impex Home Gym PM 4500 User Manual