HP Hewlett Packard Network Card 5992 4755 User Manual

HP OSMS: MySQL Sizing Guide for HP  
ProLiant c-Class Blade Servers  
HP Part Number: 5992-4755  
Published: March 2008  
Edition: 1.0  
 
Table of Contents  
Table of Contents  
3
 
4
 
6
 
8
 
Introduction  
HP Open Source Middleware Stacks (OSMS) offer building block applications, such as the Web  
Server; technical blueprints, and documents such as this MySQL Sizing Guide which describes  
the maximum MySQL workload that HP ProLiant c-Class Blade Servers are capable of supporting;  
and consulting services to speed the successful implementation of an open source strategy.  
Executive Summary  
The HP Open Source Integrated Portfolio (HP OSIP) comprises a range of products and services  
designed to verify that customers can successfully realize the cost and feature benefits of adopting  
open source software in their IT environments. HP Open Source foundation components include  
the base components of an open source-based ecosystem. HP servers and storage are validated  
to run the Linux operating system together with the supported commercial Linux distribution.  
This results published in this document were achieved using the SysBench benchmark for each  
of the HP ProLiant c-Class Blade Servers. Various test configurations and test environments were  
used and documented to provide readers an opportunity to understand the testing methodology.  
Intended Audience  
The intended audience for this document is anyone who is interested in determining the number  
of database workload that can be supported on a given HP ProLiant c-Class Blade Server and  
details for deploying a hardware environment based on the database requirements of certain  
applications.  
Scope and Purpose  
This document presents the results of a series of benchmark tests performed using the SysBench  
software test suite. The tests were conducted running a MySQL database server to evaluate  
system performance when running a database under a heavy load. The following HP ProLiant  
servers were used in this test: BL460c, BL480c, BL465c, and BL685c. The benchmark data provided  
in this sizing guide can assist customers in determining which HP Proliant c-Class Blade Server  
meets their MySQL database requirements along with how-to information for configuring the  
HP Proliant c-Class Blade Server and storage in a specific production environment.  
HP provides quality assurance from extensive integration testing with open source software and  
HP hardware so that you can confidently deploy the complete stack. Once you have completed  
a successful evaluation, you have the flexibility to “do it yourself” or get assistance from HP to  
incorporate open source stacks into your existing IT infrastructure.  
HP Services  
HP Open Source Consulting Services can help you build and integrate open source and commercial  
software across multiple operating system (OS) environments. Additionally, HP Open Source  
Support Services provide industry leading technical support for all the products HP sells,  
including hardware, operating systems, and open source middleware.  
To learn more about HP Open Source Consulting and Support Services, contact your local HP  
sales representative or visit the HP Business and IT Services website at:  
For the location of the nearest sales office, call:  
In the United States: +1 800 637 7740  
In Canada: +1 905 206 4725  
In Japan: +81 3 3331 6111  
In Latin America: +1 305 267 4220  
In Australia/New Zealand: +61 3 9272 2895  
Introduction  
9
 
         
In Asia Pacific: +8522 599 7777  
In Europe/Africa/Middle East: +41 22 780 81 11  
10  
 
Typographic Conventions  
This document uses the following typographic conventions.  
Command  
ComputerOut  
Ctrl-x  
A command name or qualified command phrase.  
Text displayed by the computer.  
A key sequence. A sequence such as Ctrl-x indicates that  
you must hold down the key labeled Ctrl while you press  
another key or button.  
ENVIRONVAR  
The name of an environment variable, for example, PATH.  
[ERRORNAME]  
The name of an error, usually returned in the errno  
variable.  
Key  
The name of a keyboard key. Return and Enter both refer  
to the same key.  
Term  
UserInput  
The defined use of an important word or phrase.  
Commands and other text that you type.  
VARIABLE  
The name of a placeholder in a command, function, or  
other syntax display that you replace with an actual value.  
\ (continuation character) A backslash (\) at the end of a line of code (such as a  
command) indicates that the following line of code is  
contiguous, and you must not insert a line break. This  
convention facilitates the typesetting of long lines of code  
examples on a printed page. If you cut and paste sample  
code from this publication, ensure that you remove  
backslash characters at line endings.  
...  
The preceding element can be repeated an arbitrary number  
of times.  
|
Separates items in a list of choices.  
Publishing History  
The document publishing date and part number indicate the current edition of the document.  
The publishing date changes when a new edition is printed. Minor changes might be made  
without changing the publishing date. The document part number changes only when extensive  
changes are made. Document updates might be issued between editions to correct errors or  
document product changes. For the latest version of this document online, see the HP Technical  
Documentation website at:  
Manufacturing Part  
Number  
Supported OS  
Edition  
Number  
Publication Date  
5992-4755  
Red Hat Enterprise Linux 5 AS Update 1  
(RHEL5ASu1)  
1
March 2008  
HP Encourages Your Comments  
HP encourages your comments concerning this document. We are committed to providing  
documentation that meets your needs. Send any errors found, suggestions for improvement, or  
compliments to:  
Include the document title, manufacturing part number, and any comment, error found, or  
suggestion for improvement you have concerning this document.  
Introduction  
11  
 
     
Benchmark Software  
The SysBench test suite is an open source, cross-platform, multi-threaded benchmark tool for  
evaluating system performance when running a database under a heavy load. Depending on  
the options used, SysBench can be used to test the following system components:  
File I/O performance  
Scheduler performance  
Memory allocation and transfer speed  
POSIX threads implementation performance  
Database server performance (OLTP benchmark)  
The results documented in this document were obtained using the SysBench test suite to generate  
results in an environment using the MySQL database server on Proliant systems (x86_64) running  
RHEL. Table 1 lists the software versions used for the benchmark tests.  
Table 1 Test Software Environment  
Software  
Website  
SysBench version 0.4.8  
Red Hat Enterprise Linux 5 AS Update 1 (RHEL5 AS u1) N/A  
MySQL database server version 5.0.46 (for an x86_64  
system)  
This paper describes the results of overall database server performance based on the OLTP  
benchmark. Because this represents a transactional workload, the InnoDB storage engine was  
used for the MySQL database server. Four different servers from the HP ProLiant family were  
used for each of the OLTP tests. Each test was performed using read-only queries and then with  
read/write queries and a varied number of connections. Additionally, each of the tests were  
performed when the MySQL data files were located on the local attached storage (SAS), as well  
as Fibre Channel attached Storage Area Network (SAN) on an HP StorageWorks 1500 Modular  
Smart Array (MSA1500).  
Hardware Test Environment  
This benchmark was performed within an HP BladeSystem c-Class blade enclosure with enhanced  
backplane using C-GbE2 Network interconnect switches. Table 2 lists the specifications of the  
hardware test environment.  
Table 2 Test Hardware Environment  
Hardware  
BL460c  
BL465c  
BL480c  
BL685c  
CPU  
(2) Quad-Core Intel® (2) Dual-Core AMD  
Xeon® processor X5365 Opteron™ processor  
(2) Quad-Core Intel® (2) Dual-Core AMD  
Xeon® processor X5365 Opteron processor  
(3.0 GHz/4MB L2 cache 2218 (2.6 GHz/1MB L2 (3.0 GHz/4MB L2 cache Model 8218 (2.6  
per core)  
cache per core)  
per core)  
GHz/1MB L2 cache per  
core)  
Memory  
16GB (8x2GB PC2-5300 14GB (6x2GB+ 2x1GB 24GB (12x2GB  
12GB (6x2GB PC2-5300  
DDR2)  
DDR2)  
PC2-5300 DDR2)  
PC2-5300 DDR2)  
Disk Storage  
2 x 73GB SAS Smart  
2 x 73GB SAS Smart  
4 x 73GB SAS Smart  
2 x 73GB SAS Smart  
Array E200i controller Array E200i controller Array P400i controller Array E200i controller  
with Battery Backed  
Write Cache (BBWC)  
and Fibre Channel  
attached MSA1500  
with Battery Backed  
Write Cache (BBWC)  
and Fibre Channel  
attached MSA1500  
with Battery Backed  
Write Cache (BBWC)  
and Fibre Channel  
attached MSA1500  
with Battery Backed  
Write Cache (BBWC)  
and Fibre Channel  
attached MSA1500  
12  
 
       
Table 2 Test Hardware Environment (continued)  
Hardware  
BL460c  
BL465c  
BL480c  
BL685c  
File System  
Partitioning  
(ext3 – no LVM):  
Disk0: /, /boot, swap  
(ext3 – no LVM):  
Disk0: /, /boot, swap  
(ext3 – no LVM):  
Disk0: /, /boot, swap  
(ext3 – no LVM):  
Disk0: /, /boot, swap  
Disk1: FC attach – 7  
Disk1: FC attach – 7  
Disk1: FC attach – 7  
Disk1: FC attach – 7  
36GB disks 10k RPM, 36GB disks 10k RPM, 36GB disks 10k RPM, 36GB disks 10k RPM,  
RAID 0  
RAID 0  
RAID 0  
RAID 0  
Operating  
System  
RHEL5 AS u1 SMP  
kernel (2.6.18-53.el5)  
with errata updates  
RHEL5 AS u1 SMP  
kernel (2.6.18-53.el5)  
with errata updates  
RHEL5 AS u1 SMP  
kernel (2.6.18-53.el5)  
with errata updates  
RHEL5 AS u1 SMP  
kernel (2.6.18-53.el5)  
with errata updates  
SysBench Configuration  
The OLTP test mode emulates real database usage by issuing common SQL queries on typical  
table structures. SysBench populates the following table with oltp-table-size rows at the beginning  
of the test.  
CREATE TABLE `sbtest` (  
`id` int(10) unsigned NOT NULL auto_increment,  
`k` int(10) unsigned NOT NULL default '0',  
`c` char(120) NOT NULL default '',  
`pad` char(60) NOT NULL default '',  
PRIMARY KEY (`id`),  
KEY `k` (`k`)  
) ENGINE=InnoDB DEFAULT CHARSET=latin1  
To test general purpose database performance, the complex test mode was used. The InnoDB  
storage engine supports ACID compliant transactions; therefore, the SysBench uses  
BEGIN/COMMITstatements around each of the transactions. Table 3 lists the SQL queries used  
in testing.  
Table 3 SQL Queries  
Type of Query  
Point queries  
Range queries  
SQL Query Syntax  
SELECT c FROM sbtest WHERE id=N  
SELECT c FROM sbtest WHERE id BETWEEN N and M  
SELECT SUM(c) FROM sbtest WHERE id BETWEEN N AND M  
Range SUM()  
queries  
Ordered range  
queries  
SELECT c FROM sbtest WHERE id BETWEEN N AND M ORDER BY c  
SELECT DISTINCT c FROM sbtest WHERE id BETWEEN N AND M ORDER BY c  
UPDATE sbtest SET k=k+1 WHERE id=N  
Distinct range  
queries  
Updates on  
indexed column  
Updates on  
non-index column  
UPDATE sbtest SET c=N WHERE id=N  
Delete queries  
Insert queries  
DELETE FROM sbtest WHERE id=N  
INSERT INTO sbtest VALUES (...)  
Table 4 lists the configuration options that were used for SysBench.  
SysBench Configuration  
13  
 
   
Table 4 SysBench Options  
SysBench option  
Value  
Description  
--test  
oltp  
OLTP emulates operations that a read database would perform. Other  
options are CPU, threads, mutex, memory, and fileio.  
--oltp-table-size  
1000000  
Sets the number of rows in the test table.  
--mysql-engine-trx yes  
Sets the value for whether the storage engine used is transactional.  
--oltp-test-mode  
complex  
Determines how transactions are performed. Options are simple (very  
simple SELECTs), complex (transactional), and nontrx (non-transactional).  
--oltp-read-only .  
on/off  
Sets whether the UPDATE/DELETE/INSERT queries are used. Both cases  
were run for these tests.  
--oltp-dist-type  
--max-requests  
special  
0
Allows for the ditribution of random numbers  
Sets the total number of requests allowed. The value “0” indicates  
unlimited requests are allowed.  
--num-threads  
--init-rng  
varied from 1 to 128 Sets the number of worker threads.  
on  
Initializes the random number generator.  
--max-time  
180  
Sets the limit for total execution time in seconds.  
MySQL Installation  
The MySQL Certified Software, downloaded from the MySQL Network, provides the  
mysql-enterprise-gpl-5.0.46-0.rhel5.x86_64.tarfile for RHEL5 on x86_64 servers.  
Table 5 lists the components contained in the tar file.  
Table 5 MySQL Installation Packages  
RPM Package  
Description  
MySQL-server-enterprise-gpl-5.0.46-0.rhel5.x86_64.rpm  
MySQL-client-enterprise-gpl-5.0.46-0.rhel5.x86_64.rpm  
MySQL database server (mysqld, etc)  
MySQL client programs (mysql,  
mysqladmin, etc)  
MySQL-devel-enterprise-gpl-5.0.46-0.rhel5.x86_64.rpm  
MySQL header files and libraries  
(needed to build SysBench )  
MySQL Configuration  
The example configuration file (my-large.cnf) that is provided with the MySQL database  
server, was used as the basis of the my.cnftest file. Table 6 lists the changes that were made to  
the mysqld configuration.  
Table 6 MySQL Test Configuration  
mysqld Option  
Value  
1600  
10  
Description  
max_connections  
max_connect_errors  
Maximum number of connections. The default is 100.  
Number of interrupted connections from a host after which  
connections are blocked.  
table_cache  
2048  
Number of open tables for all threads. The default is 512.  
sort_buffer_size  
512K  
Buffer size for each thread that needs to sort results. Increase  
for faster ORDER BY and GROUP BY operations.  
binlog_cache_size  
1M  
Size of cache to hold SQL statements for the binary log  
during a transaction. Increase for large, multiple-statement  
transactions.  
14  
 
         
Table 6 MySQL Test Configuration (continued)  
mysqld Option  
Value  
Description  
join_buffer_size  
1M  
Buffer size for joins that do not use indices and require a  
full table scan.  
thread_cache_size  
thread_stack  
16  
Number of server threads to cache for reuse.  
Per-thread stack size. The default is 192K.  
192K  
0
query_cache_size  
Amount of memory allocated for caching query results. The  
value “0” disables query cache.  
ft_min_word_len  
4
Minimum length of the word to be included in a full-text  
index.  
tmp_table_size  
64M  
Maximum size of in-memory temp tables. Increase for  
GROUP BY.  
innodb_data_file_path  
innodb_buffer_pool_size  
ibdata1:100M\  
:autoextend  
Path to innodb data files and sizes. Increase size from 10MB  
to 100MB.  
1900M  
Number of bytes of the memory buffer to cache data and  
indices.  
innodb_additional_mem_pool_size 20M  
Number of bytes of memory pool used to store the data  
dictionary information and data structures. The default is  
1M.  
innodb_log_file_size  
900M  
Size of each log file in log group. The default is 5MB.  
innodb_flush_log_at_trx_commit 2  
Determines how the log buffer is written to the disk. The  
value “2” means data is flushed on every commit, but no  
disk flush occurs. The default is 1.  
innodb_locks_unsafe_for_binlog 1  
Controls next-key locking in search and index scans. The  
default is 0 (disabled).  
innodb_thread_concurrency  
innodb_concurrency_tickets  
Innodb_log_buffer_size  
1000  
Number of threads currently inside the innodb engine. The  
default varies with the MySQL version.  
500  
8M  
Number of 'free tickets' to allow re-entry into the innodb  
engine.  
The size of buffer InnoDB used in the memory to write the  
log files. Ranges from 1MB to 8MB. Large log buffer saves  
disk I/O operations.  
innodb_table_locks  
innodb_support_xa  
innodb_doublewrite  
0
0
Specifies whether InnoDB locks a table internally. The  
default is 1 (enabled).  
Specifies whether InnoDB supports XA transactions. The  
default is 1 (supported).  
skip-innodb-doublewrite InnoDB stores all data twice by default. Disable it by using  
the --skip-innodb-doublewrite parameter when starting the  
MySQL server.  
innodb_checksums  
skip-innodb-checksums Checksum validation is used on all data pages by default.  
Disable it by using the --skip-innodb-checksums parameter  
when starting the MySQL server.  
Benchmark Results  
For each read-only or read/write test, SysBench was run with a varying number of thread  
connections to determine the maximum transactions per second (tps). The tests were conducted  
on both the internal storage with the battery backed write cache option (BBWC) and the Fibre  
Channel attached Storage Area Network.  
Benchmark Results  
15  
 
 
Depending on whether the application requires many read/write queries, it may help to set the  
database connection pool to the number of processing cores. For read-only testing, all HP Proliant  
c-Class Blade Servers in the test can still handle almost the same number of transactions even  
though the number of concurrent threads is double the number of processing cores. Therefore,  
more database connections than the number of processing cores for read-only applications can  
be used.  
NOTE: For read/write testing, the maximum performance (shown in bold in the following  
result tables) was achieved when the number of concurrent threads was the same as the total  
number of processing cores in the system.  
Interpreting the Data  
The results of the testing are presented in the following sections and show the maximum number  
of tps that each of the systems can handle, depending on the number of threads. The precise  
database query load and mix of queries that an application generates vary greatly between  
applications. Another important consideration in the benchmark testing was how many concurrent  
connections the application had with the database server. The results in the data tables show  
that having more connections can negatively impact the throughput.  
If an application requires 1200 read/write tps from 16 database connections, all of the HP Proliant  
c-Class Blade Servers used in these tests are suitable because all of the test servers can handle at  
least 1245 tps. For read-only applications that require 2000 tps using eight database connections,  
both the HP ProLiant BL465c and BL685c servers, configured with 2 Dual-Core Opteron  
processors, are suitable having a capacity of 2500 tps. The ProLiant BL460c server and BL480c  
server, configured with 2 Quad-Core Xeon processors, can handle a maximum of 4500 read-only  
tps and 3000 read/write tps. For an application that requires more than 4500 read-only or more  
than 3000 read/write tps, consideration should be given to using MySQL Replication and spreading  
the load across multiple ProLiant servers.  
HP Proliant BL460c Server  
The HP ProLiant BL460c server has features that are equal to standard 1U rack-mount servers.  
The two-processor, dual-core or quad-core ProLiant BL460c combines power-efficient compute  
power, and high density with expanded memory and I/O for maximum performance. The  
ProLiant c7000 enclosure supports up to 16 BL460c server blades, two more servers than the IBM  
BladeCenter, and each BL460c supports double the memory capacity of the HS21 server without  
an expansion blade. The Proliant BL460c now comes with dual-core and quad-core Intel® Xeon®  
processors, DDR2 fully buffered DIMMs, Serial Attached SCSI (SAS) or Serial ATA (SATA), hard  
drives, and support of multi-function NICs and multiple I/O cards. The BL460c provides a  
performance system ideal for the full range of scale-out applications. In this small form factor,  
the BL460c includes more features to enable high-availability, such as hot plug hard drives,  
mirrored memory, online spare memory, memory interleaving, embedded RAID capability, and  
enhanced remote Lights-Out management. Table 7 provides the benchmark results that were  
obtained using a ProLiant BL460c server configured with two Quad-Core Intel® Xeon® processors  
running at 3.0GHz.  
NOTE: Maximum performance appears in bold.  
Table 7 SysBench Results for the HP ProLiant BL460c Server  
Read/Write (tps)  
Internal SAS  
Read Only (tps)  
Internal SAS  
Read/Write (tps) Fibre  
Channel SAN  
Read Only (tps) Fibre  
Channel SAN  
Connections  
1
2
624  
794  
610  
809  
1057  
1454  
1029  
1330  
16  
 
     
Table 7 SysBench Results for the HP ProLiant BL460c Server (continued)  
Read/Write (tps)  
Internal SAS  
Read Only (tps)  
Internal SAS  
Read/Write (tps) Fibre  
Channel SAN  
Read Only (tps) Fibre  
Channel SAN  
Connections  
4
8
1893  
3088  
2604  
2382  
2220  
1478  
2615  
4301  
4374  
4148  
4006  
3641  
1910  
3145  
2690  
2423  
2271  
1626  
2593  
4297  
4353  
4271  
3990  
3722  
16  
32  
64  
128  
Figure 1 SysBench Results for the HP ProLiant BL460c Server  
HP Proliant BL465c Server  
The HP ProLiant BL465c server is a two-way server blade designed for maximum compute  
density and power efficiencies. The small form factor accommodates two blades in each slot  
allowing for 16 servers in a standard 6u enclosure. Each CPU is a 95 Watt version of the Dual-Core  
AMD Opteron 2000 series processor, which gives the equivalent of four processing units per  
server. Up to 32GB of PC2-5300 DIMMs running at 667MHz is supported. Internal storage consists  
of two small form factor disk drives, either SAS or SATA attached to the integrated Smart Array  
E200i RAID controller with 64MB cache supporting RAID 0, 1. An optional StorageWorks SB40c  
storage blade provides connectivity to additional storage, supporting RAID 5, and ADG. The  
benchmark results in Table 8 were obtained with a ProLiant BL465c server configured with two  
Dual-Core Opteron processors running at 2.6 GHz.  
Table 8 SysBench Results for the HP ProLiant BL465c Server  
Read/Write (tps),  
Internal SAS  
Read Only (tps), Internal Read/Write (tps), Fibre Read Only (tps), Fibre  
Connections  
SAS  
Channel SAN  
Channel SAN  
1
2
497  
931  
664  
514  
746  
1340  
2341  
2455  
2437  
2302  
973  
1415  
4
1665  
1467  
1245  
1014  
1692  
2568  
8
1523  
2640  
16  
32  
1298  
2621  
1048  
2445  
Benchmark Results  
17  
 
     
Table 8 SysBench Results for the HP ProLiant BL465c Server (continued)  
Read/Write (tps),  
Internal SAS  
Read Only (tps), Internal Read/Write (tps), Fibre Read Only (tps), Fibre  
Connections  
SAS  
Channel SAN  
Channel SAN  
64  
885  
621  
2097  
1962  
911  
630  
2148  
128  
2049  
Figure 2 SysBench Results for the HP ProLiant BL465c Server  
HP Proliant BL480c Server  
The HP ProLiant BL480c server is a 2-way server blade designed for maximum performance and  
scalability. The ProLiant BL480c server can support two Quad-Core Inter Xeon processors, which  
gives the equivalent of eight processing units per server. Up to 48GB (PC2-5300 DDR2, 667 MHz)  
of memory is supported. Each full-height blade has a Smart Array P400i Controller that supports  
four hot-swap SCSI disks in a RAID configuration. An optional dual-port Fiber Channel adapter  
provides connectivity to HP and third-party storage area networks (SANs). The benchmark  
results in Table 9 were obtained with a BL480c server configured with two Quad-Core Inter Xeon  
processors running at 3.0 GHz.  
Table 9 SysBench Results for the HP ProLiant BL480c Server  
Read/Write (tps),  
Internal SAS  
Read Only (tps), Internal Read/Write (tps), Fibre Read Only (tps), Fibre  
Connections  
SAS  
Channel SAN  
Channel SAN  
1
2
628  
817  
650  
944  
1060  
1922  
3125  
2663  
2411  
2261  
1604  
1494  
2939  
4544  
4635  
4533  
4260  
3926  
1070  
1573  
4
1951  
2987  
8
3177  
4588  
16  
32  
64  
128  
2655  
4682  
2435  
4551  
2251  
4333  
1634  
3986  
18  
 
     
Figure 3 SysBench Results for the HP ProLiant BL480c Server  
HP Proliant BL685c Server  
The HP ProLiant BL685c server blade delivers no-compromise performance and expansion in  
the densest four processor server blade form factor available. With up to four AMD Opteron™  
8000 Series processors, 64GB of DDR2 memory, two hot-plug Serial Attached SCSI (SAS) or Serial  
ATA (SATA) hard-drives, four embedded Gigabit NICs, and three I/O expansion slots, the HP  
ProLiant BL685c delivers the density you want with the performance you need to handle the  
most demanding enterprise class applications. Each blade server has a Smart Array E200i  
Controller that supports two hot-swap SCSI disks in a RAID configuration. Optional dual-port  
Fiber Channel adapter provides connectivity to HP and third-party storage area networks (SANs).  
The benchmark results are listed in Table 10, and were obtained using a ProLiant BL685c server  
configured with two Dual-Core Opteron processors running at 2.6 GHz.  
Table 10 SysBench Results for the HP ProLiant BL685c Server  
Read/Write (tps),  
Internal SAS  
Read Only (tps), Internal Read/Write (tps), Fibre Read Only (tps), Fibre  
Connections  
SAS  
Channel SAN  
Channel SAN  
1
2
470  
951  
670  
497  
664  
1307  
2327  
2419  
2400  
2312  
2094  
1971  
984  
1323  
4
1650  
1540  
1463  
1385  
1247  
1034  
1770  
2426  
8
1655  
2427  
16  
32  
64  
128  
1548  
2413  
1468  
2347  
1319  
2139  
1050  
1975  
Benchmark Results  
19  
 
     
Figure 4 SysBench Results for the HP ProLiant BL685c Server  
Summary  
The results of the SysBench testing indicated that the HP Proliant c-Class Blade Servers provide  
a solid platform for the MySQL database server. Along with the integrated Smart Array controller  
these servers provide the security of RAID storage to the host operating system and database  
files, while also offering excellent performance for many applications.  
The results of the benchmarks tests are provided in Figure 5 and indicate that the HP ProLiant  
BL465c server is a good choice for an entry-level database server with scalability to SAN, as the  
size of the database grows. When higher performance levels are required from a single server,  
both the HP ProLiant BL460c and HP ProLiant BL480c servers (when configured with two  
Quad-Core processors) provided a good option for scalability.  
Figure 5 Benchmark Test Results  
The test results indicated that optimum price and performance could be achieved using a MySQL  
Replication configuration with a pair of HP ProLiant BL460c or BL480c servers with 8 cores in a  
high-availability cluster where HP Proliant c-Class Blade Servers acted as the MySQL master  
database server. This, combined with a number of ProLiant BL465c servers acting as the MySQL  
slave database servers, provided a robust, scalable database environment within a single HP  
BladeSystem enclosure.  
Resources  
For additional information on the hardware and software used in the MySQL sizing tests, see  
the following websites:  
20  
 
       

Grizzly Sander H8253 User Manual
Grizzly Saw G8993 User Manual
Grundig Boat MM 8050 User Manual
Haan Vacuum Cleaner BS 20 User Manual
Harmar Mobility Personal Lift 650 User Manual
Heath Zenith Home Security System CB 2011 User Manual
HotBrick Network Card VPN Client User Manual
HP Hewlett Packard Laptop tx1138ea User Manual
HTC Cell Phone Jetstream User Manual
IBM Portable Media Storage SAN384B User Manual