sid meier s civilization 3 free download full versionpowerdvd ultra 9 free downloadnero backitup 4 free downloadpartition magic 7 download full version
The CentOS Project is really a community-driven free software effort devoted to delivering a strong open source ecosystem. For users, currently a consistent manageable platform that meets a wide variety of deployments. For open source communities, currently a solid, predictable base to make upon, in addition to extensive resources to construct, test, release, and keep their code.
We re expanding the production of CentOS Linux images across many vendors, providing official images for Amazon, Google, plus more. For self-hosted cloud, we also give a generic cloud-init enabled image.
For more details about updates and improvements in CentOS Linux 7, take a look at the release notes or the making announcement within the mailing list archive.
CentOS Dojo - Brussels, Belgium Stop from the CentOS Booth at Fosdem, and turn into for the devroom talks
CentOS couldn't survive possible without worrying about support in our sponsors. We would like to thank the next product/service if you are a CentOS sponsor:
The following procedure is usually a step-by-step guide Cookbook with tips and data for installing Oracle Database 10g on Red Hat Linux.
See also Oracle Database Release Notes 10g Release 2 10.2 for Linux x86
Red Hat Enterprise Linux Advanced Server 2.1
These instructions may also work for Red Hat Enterprise Linux ES 2.1, 3, and 4.
Works much like on RHEL AS 3.
Mandrake 10.0 Community Edition
Compute a cyclic redundancy check CRC checksum with the downloaded files and compare the checksum numbers from the numbers posted on OTNs website. For example:
cpio - idmv /10.1.0.3.0/1 /10.1.0.3.0 /10.1.0.3.0/1/DataFiles /10.1.0.3.0/1 /10.1.0.3.0
mkisofs - r Disk1 cdrecord - v dev0, 0, 0 speed20 -
Drives speed varies; you will get the dev numbers if you execute cdrecord - scanbus.
Note that 10g R2 wont fit for a passing fancy CD as it has over 780MB.
Oracle says how the system need to have at least 512MB of RAM and 1GB of swap space or twice the dimensions of RAM. And for systems with 2 GB of RAM, the swap space might be between one as well as times the length of RAM. You might also need to check out Sizing Swap Space.
For test sake I tried to set up an Oracle Database 10g Type: General Purpose Database using a little PC with 256MB of RAM and 1 GB of swap space. I was able to find a 10g database operational on this little PC without having a problem.
To check the dimensions of physical memory, execute:
To check how big is swap space, execute:
You may also add temporary swap space for your system by setting up a temporary swap file rather than using a raw device. Here could be the procedure:
According to Oracles documentation, the Oracle Universal Installer OUI requires around 400 MB of free space from the
directory. But OUI checks if
is only over 80 MB.
filesystem, you'll be able to temporarily create a
directory in another filesystem. Here is how you are able to do this:
When you're done together with the Oracle installation, shutdown Oracle and take off the temporary
Before you install an Oracle Database 10g you need to look into the system for required RPMs. On my systems I usually buy a minimum report on RPMs which normally requires the installing additional packages for Oracle databases. Always ensure make use of the latest RPMs and kernels!
4 compat-db-4.1.25-9 control-center-2.8.0-12 4 4 glibc-2.3.4-2 glibc-common-2.3.4-2 gnome-libs-1.4.1.2.90-44.1 4 4 make-3.80-5 pdksh-5.2.14-30 sysstat-5.0.5-1 4.2
RPM or possibly a newer version! Otherwise the OUI prerequisite check will fail.
To check if you might be running the x8664 kernel over a x8664 platform, run:
It is essential to have these x8664 RPMs installed. The above command are listed the architecture of the binary package. You will see that some RPMs are installed twice after you run this command x86 RPM and x8664 RPM. You need to make sure that all required x86-64 RPMs listed below are installed.
4 compat-db-4.1.25-9 control-center-2.8.0-12 4 4 glibc-2.3.4-2 glibc-common-2.3.4-2 gnome-libs-1.4.1.2.90-44.1 4 4 make-3.80-5 pdksh-5.2.14-30 sysstat-5.0.5-1 4.2
RPM or even a newer version! Otherwise the OUI prerequisite check will fail.
make-3.79.1 gcc-3.2.3-34 glibc-2.3.2-95.20 compat-db-4.0.14-5 compat-gcc-7.3-2.96.128 compat-gcc-c-7.3-2.96.128 compat-libstdc-7.3-2.96.128 compat-libstdc-devel-7.3-2.96.128 openmotif21-2.1.30-8 setarch-1.3-1
RPM or perhaps a newer version! Otherwise the OUI prerequisite check will fail.
make-3.79.1 gcc-3.2.3-34 glibc-2.3.2-95.20 glibc-devel-2.3.2-95.20 glibc-devel-2.3.2-95.20 32 bit compat-db-4.0.14-5 compat-gcc-7.3-2.96.128 compat-gcc-c-7.3-2.96.128 compat-libstdc-7.3-2.96.128 compat-libstdc-devel-7.3-2.96.128 gnome-libs-1.4.1.2.90-34.1 32 bit openmotif21-2.1.30-8 setarch-1.3-1 libaio-0.3.96-3 libaio-devel-0.3.96-3
To check if that you are running the x8664 kernel on the x8664 platform, run:
It is essential to have the right x86 and x8664 RPMs installed. The above command can place the architecture of the binary package. And as you are able to see inside the above list,
and also other RPMs are listed twice. This means you have to setup packages for both architectures, x86 and x8664.
make-3.79.1 gcc-3.2.3-34 glibc-2.3.2-95.20 compat-db-4.0.14-5 compat-gcc-7.3-2.96.128 compat-gcc-c-7.3-2.96.128 compat-libstdc-7.3-2.96.128 compat-libstdc-devel-7.3-2.96.128 openmotif21-2.1.30-8 setarch-1.3-1
make-3.79.1 glibc-2.2.4-32 gcc-2.96-128 gcc-c-2.96-128 libstdc-2.96-128 openmotif-2.1.30-11
However, since its listed being a requirement for RHEL 3 Id recommend to set up it on RHEL3 and so on newer releases at the same time. Its not a desire for RHEL 2.1 though.
For RHEL 3 and RHEL 2.1 it is usually important to have
or even a newer version installed. Make sure you've got the
package. Also, Red Hat changed the version naming schema from
in RHEL3 Update 5. This appears to confuse OUI in RHEL3 U5 because it complaining which it cant discover the right
version. I ignored this. The
which worked fine and I didnt get problems. To confirm the RPM, run:
package is installed. Earlier versions of RHEL 3 didnt do the installation by default if you selected a baseline system installation:
utility is completely in RHEL4, RHEL3 and Fedora Core. It is utilized to tell the kernel to report some other architecture as opposed to current one. It is also familiar with emulate a 3GB virtual address space for applications which don't run properly which has a larger virtual address space. To look at the RPM, run:
Note that you simply also need to install this i386 and x8664 RPMs or even already installed, or perhaps you will get various error messages.
If you havent installed Update 3 or later, dont forget to set up an updated
RPM from Oracle or RHN, you will get the next error message:
usr/bin/ld:/u01/app/oracle/oracle/product/10.2.0/db1/lib//libirc.afastmemcpy.o: relocation RX8664PC32 against memcpymemopsmethod cannot be usedwhen building a shared object; recompile with - fPIC/usr/bin/ld: final link failed: Bad value collect2: ld returned 1 exit status
RPM are listed as requirements. On my system I failed to install these RPMs since Im against installing desktop stuff on servers, see list below. When I installed 10g R2 I would not experience any problems when these RPMs were missing. But just for completeness Ive listed each of the RPMs you may have to set up if you wish to install the
RPM are listed as requirements. On my system I failed to install these RPMs since Im against installing desktop stuff on servers. When I installed 10g R2 I would not experience any problems when these RPMs were missing. When you wish to install
then have some fun. The list might be very long with each of the dependencies like
NOTE: No certification continues to be projected for 10g R1 64-bit on RHEL AS 4 x8664 before this writing.
You could possibly get various different error messages throughout the installation otherwise all of the RPMs were installed on your system for 10g R1 64-bit.
Note the 10g 10.1.0.3 OUI Product-specific Prerequisite check will fail for that
versions. You can ignore these failed checks and proceed.
RPM should already be installed automagically. But remember that 10.1.0.3.0 OUI doesn't recognize RHEL AS 4 being a supported release yet. This means you must edit the
file, see below, or maybe you apply the 4153257 patch for 10g R1 on RHEL 4. 10g R2 does recognise RHEL AS 4 like a supported platform.
Note, if you do not install the i386
RPM, you are certain to get an error message such as this one:
On my RHEL AS 2.1 x86 system I had to install this RPMs and dependencies to meet up with the software requirements:
in order to secure Oracles Product-specific Prerequisite checks. Oracles recommended
version is 2.2.4.31.7 or more.
Also, its important to set up a newer kernel version for RHEL AS 2.1. Definitely dont employ a kernel more than 2.4.9-e.25. To examine the kernel version run
version that accompany FC3 is not going to pass Oracles Product-specific Prerequisite checks. I ignored this warning on FC3.
FC1 will not be supported by Oracle you could a hard time to obtain the right RPM version for Oracle10g. I tried to make use of the RPMs that had the FC1 CDs that are easily accessible to everyone. I would not had any problems installing an Oracle Database 10g General Purpose Database on FC1 with one of these RPMs.
RPM for RH9. RH9 is just not supported by Oracle.
This RPM is vital for RHEL since RHEL AS 4, RHEL AS 3, and RHEL AS 2.1 are Linux releases based on Oracle. Without this RPM, Oracle 10g OUI wont be competent to recognize it to be a supported OS. However, the installer of 10g 10.1.0.3 isn't going to recognize RHEL AS 4 to be a supported release yet. This means that you just will have to edit the
file. You also have to change
for unsupported releases like Fedora Core and RH9.
since this can disable other checks maybe you dont desire to.
file to produce Oracle 10g believe its running with a supported release.
Regarding RHEL AS 4, the installer for 10g 10.1.0.3 won't recognize RHEL AS 4 to be a supported release but 10g R2 OUI does.
file, you'll be able to simply copy/paste this commands:
After you're done with all the Oracle10g installation, undo adjustments you made to
since the OUI isn't very clever. If OUI finds digit number 3 inside
file, you pass the main system version check because it believes its a RHEL3 system.
To see all kernel parameters, execute:
For Oracle10g, this kernel parameters need to be set to values over or similar to the recommended values which could be changed from the
shmmax 2147483648 To verify, execute: cat/proc/sys/kernel/shmmax shmmni 4096 To verify, execute: cat/proc/sys/kernel/shmmni shmall 2097152 To verify, execute: cat/proc/sys/kernel/shmall for 10g R1 shmmin 1 To verify, execute: ipcs - lm grep min seg size shmseg 10 Its hardcoded inside the kernel - the default is significantly higher semmsl 250 To verify, execute: cat/proc/sys/kernel/sem awk print 1 semmns 32000 To verify, execute: cat/proc/sys/kernel/sem awk print 2 semopm 100 To verify, execute: cat/proc/sys/kernel/sem awk print 3 semmni 128 To verify, execute: cat/proc/sys/kernel/sem awk print 4 file-max 65536 To verify, execute: cat/proc/sys/fs/file-max iplocalportrange 1024 65000 To verify, execute: cat/proc/sys/net/ipv4/iplocalportrange
NOTE: Do not customize the value of any kernel parameter using a system where it can be already more than listed as minimum requirement.
to meet up with the minimum requirement. On RHEL AS 4 x8664 I had to increase
for outgoing messages to 1024 65000 that is needed for high-usage systems. This kernel parameter defines your neighborhood port range for TCP and UDP people to choose from.
But to ensure that these new added lines or settings in
become effective immediately, execute these command:
Starting with 10g R2 some network settings should be adjusted too which is checked by OUI. For details on setting these parameters, see Changing Network Kernel Settings.
For more details on shmmax, shmmni, shmmin, shmseg, and shmall, see Setting Shared Memory.
For more details on semmsl, semmni, semmns, and semopm, see Setting Semaphores.
Oracle says that about 2.5 GB of disk space ought to be reserved to the Oracle software on Linux.
When I did an Oracle 10g Release 1 10.1.0.3.0 General Purpose Database installation not including any software from your Oracle Database 10g Companion CD, the Oracle software used about 1.3 GB of disk space, plus the preconfigured General Purpose Database datafiles, etc. used about 710 MB of disk space.
If moreover, you may install additional software on the Oracle Database 10g Companion CD, then add at the least 1 GB of free disk space.
So when you install Oracle10g Enterprise Edition and other software in the Oracle Database 10g Companion CD, you will want about 2.5 GB of disk for that Oracle software. And when you also wish to add a preconfigured database on the very same filesystem, make sure you add another 1 GB of disk space.
NOTE: If you dont put Oracle10g over a separate filesystems, and then make sure the fundamental filesystem
has enough disk space. You can confirm the free space of the foundation filesystem with the next command:
account and groups, execute the subsequent commands:
Most shells like Bash provide control of various resources such as maximum allowable variety of open file descriptors or even the maximum amount of processes on the market to a user. For additional information on
If you just purchase a small test database, you then might be ok while using current settings note which the limits usually vary. But for larger production databases, you ought to increase the subsequent shell limits to this values recommended by Oracle:
nofile 65536 To verify, execute: ulimit - n nproc 16384 To verify, execute: ulimit - u
option denotes the maximum amount of open file descriptors, and
denotes the maximum amount of processes on the market to a single user.
To see all shell limits, execute:
and ways in which to improve the limit, see Setting Limits for your Maximum Number of Processes to the Oracle User. Even though this procedure was written for Oracle9i on RHAS 2.1, furthermore, it applies to Oracle10g on RHEL AS 2.1, RHEL AS 3, along with other versions.
But if you wish to comply with Oracles Optimal Flexible Architecture OFA, you then dont need to place the database files from the
This is not absolutely vital but if you would like to comply with OFA, you then might need to create the next directories likewise:
In this situation, orcl stands for that name with the database which will likely be the name from the instance. This is often the case for single instance databases.
Optimal Flexible Architecture OFA for 10g R1 10.1.0.2
The OFA standard can be a guideline produced by Oracle to guarantee reliable Oracle installations. For Oracle 10g Database, the OFA recommended Oracle home path has evolved.
The home path with the first 10g 10.1.0 database installation over a system will be:
u01/app/oracle/product/10.1.0/db1
If you would use a second Oracle 10g Database 10g 10.1.0 for a passing fancy system, the Oracle home directory could well be as follows:
u01/app/oracle/product/10.1.0/db 2
but from the user oraowner, next the path in the Oracle home directory will be:
u01/app/oraowner/product/10.1.0/db1/u01/app/oraowner/product/10.1.0/db2
u01/app/oracle/product/10.1.0/db1
etc. which complies with all the OFA guidelines. But others could be used, one example is:
disk1/app/oracle/product/10.1.0/db1
ought to be used for that Oracle software only.
etc. needs to be used for your database files. The
initialization parameter that's typically the just like the SID term for single instance databases.
account, some environment variables has to be configured because of this account before OUI commences.
is used, then Oracle recommends that you just dont set the
environment variable but that you simply choose the default path suggested because of the OUI. You can set and use
once you finished running OUI.
Also, the earth variables
should stop set. If youve already set these environment variables, you may unset them by running the next commands:
you may add these environment variables to your
file which can be the user startup file for your Bash shell on Red Hat Linux. To do this you may simply copy/paste this commands to generate these settings permanent for your
If you dont install Oracle on your own local system but with a remote server, then you'll want to relink X in your local desktop. The easiest way to make this happen is to work with the X11 forwarding feature of ssh. This means that you just dont have to perform
Here is definitely an example how to create use in the X11 forward feature of ssh. Simply run this command from a local desktop :
Now after you try to perform any GUI tool about the remote server, it'll automatically be relinked in your local desktop. If this will not be working, verify which the
however, you have to set
or change on the directory containing the image directory
If you install Oracle10g from your CD, mount the CD by running these commands in another terminal:
ensure that the Oracle environment variables are set, see Setting Oracle Environments. You can verify the settings by running the
from your mounted CD, run these command because the
Note, the screens and questions will be different when you install 10g R2 or 64-bit 10g R1 database
Welcome Screen: - Basic Installation: Checked it which can be the default - Oracle Home Location: Use default:/u01/app/oracle/product/10.1.0/db1 - Installation Type: I used the default: Enterprise Edition - UNIX DBA Group: Use default: dba - Create Starter Databases: I checked it because of this example and that is the default - Global Database Name: orcl - Database password: Type inside the password for SYS, SYSTEM, SYSMAN, and DBSNMP accounts - Advanced Installation: For this article I failed to check it Click Next - Specify Inventory directory and credentials: - Full path on the inventory directory: Use default:/u01/app/oracle/oraInventory - Specify Operating System group name: Use default: oinstall Click Next - A window pops nearly run the script: Run the script in another terminal: su - root Creating the Oracle inventory pointer file Changing groupname of/u01/app/oracle/oraInventory to oinstall. Click Continue - Product-specific Prerequisite Checks: Verify that every checks happen to be passed. Make sure which the status of the Check is scheduled to Succeeded. On RHEL AS 4 neglect the warnings for binutils, gcc, and openmotif and proceed. If a check failed, see Oracle10g Installation Errors on Linux. Note how the Retry button doesnt work when you finally fixed one with the failed checks. Click Next - Select Database Configuration: I selected General Purpose. Click Next - Specify Database Configuration Options: - Global Database Name: I used orcl. - SID: I used orcl. Click Next - Select Database Management Option: I selected Use Database Control for Database Management. Click Next - Specify Database File Storage Option: I selected File System. - File System - Specify Database file location:/u01/app/oracle/oradata/
After Oracle10g may be installed, be sure that
could possibly be different on the body!
For 10g R1 10.1.0.3 I added the next lines to your
export ORACLEHOMEORACLEBASE/product/10.1.0/db1 export PATHPATH:ORACLEHOME/bin export LDLIBRARYPATHORACLEHOME/lib
For 10g R2 10.2.0.1.0 I added this lines to your
export ORACLEHOMEORACLEBASE/oracle/product/10.2.0/db1 export PATHPATH:ORACLEHOME/bin export LDLIBRARYPATHORACLEHOME/lib
environment variable. Otherwise you is certain to get the error
The slash connects you to your schema of SYS. In the above example you'll be connected towards the schema of SYS using the privilege SYSDBA. SYSDBA gives you the next privileges:
If you installed a preconfigured database using OUI, then several Oracle background processes are running on your own server. Execute these command to find out the background processes:
To shutdown the Oracle background processes after an Oracle Database 10g installation, you may execute the subsequent commands:
During the installing Oracle 10g, OUI offered two Database Management Options:
If you selected Database Control for Database Management, then a Oracle Enterprise Manager Database Control Database Control may be shutdown with the subsequent command which stops the two agent plus the Oracle Containers for Java OC4J management service:
If you selected Grid Control for Database Management and that is used for full Grid Control installations, then your Oracle Management Agent standalone agent with the Oracle Enterprise Manager Grid Control Grid Control could be stopped with the subsequent command:
To shutdown Oracle CSS daemon, run:
To reinstall Oracle10g after having a failed installation attempt, you might would like to execute this commands.
Make sure you first of all used the De-installation option in OUI.
su - root export ORACLEHOME/u01/app/oracle/product/10.1.0/db1. ORACLEHOME/bin/localconfig delete stops the Oracle CSS daemon and deletes configuration rm - rf/u01/app/oracle rm - f/etc/oratab rm - rf/etc/oracle rm - f rm - f/usr/local/bin/coraenv/usr/local/bin/dbhome/usr/local/bin/oraenv
Here is often a list of common Oracle10g installation problems along with issues.
Note that most with the issues are set to not following correctly not hard to install procedure. And some errors are set to not employing an Oracle supported Linux OS.
Starting Oracle Universal Checking installer Checking main system version: have to be redhat-2.1, UnitedLinux-1.0 or redhat-3 Failed Exiting Oracle Universal Installer, log just for this session is usually found
Checking for gcc-2.96; found Not found. Failed
Note that Retry from the Product-specific Prerequisite Checks window won't work. So you either place it manually to Passed or perhaps you restart OUI.
Checking for openmotif-2.1.30-11; found Not found. Failed
Note that Retry from the Product-specific Prerequisite Checks window will not work. So you either set it up manually to Passed or perhaps you restart OUI.
Checking for shmmax2147483648; found shmmax33554432. Failed
Note that Retry inside Product-specific Prerequisite Checks window will not work. So you either work it manually to Passed otherwise you restart OUI.
Note that Retry within the Product-specific Prerequisite Checks window isn't going to work. So you either work it manually to Passed or perhaps you restart OUI.
Checking for filemax65536; found filemax26163. Failed
Note that Retry within the Product-specific Prerequisite Checks window doesn't work. So you either the choice is yours manually to Passed or perhaps you restart OUI.
is defined correctly, then you most likely have a trailing slashon the
environment variable. Remove it and try again to attach to sys e.g from
ORACLEHOME/u01/app/oracle/product/10.1.0/db1
ORACLEHOME/u01/app/oracle/product/10.1.0/db1
ORA-00988 missing or invalid passwords.
During the Oracle10g installation you most likely provided a password for your Oracle database accounts that started that has a digit number. Ignore this error message and modify the password after you are completed with the Oracle10g installation.
sysresv - i sysresv: error while loading shared libraries: 10.1: cannot open shared object file: No such file or directory
X11 connection rejected as a result of wrong authentication. X link to localhost:10.0 broken explicit kill or server shutdown.
To rectify this concern, attempt to login to your remote Oracle server again using the X11 forward feature of ssh. Execute these command through your local desktop :
Now whenever you try to operate any GUI tool around the remote server, it'll automatically be relinked for your local desktop. If this just isn't working, verify the
NOTE: If you use by way of example Red Hat Fedora Core 3 because your desktop and you desire to install the database on another machine, then you'll want to set the
After that you have to restart your X server. I usually do this together with the
however, you must set
ORA-00988 missing or invalid passwords.
During the Oracle10g installation you almost certainly provided a password for that Oracle database accounts that started having a digit number. Ignore this error message and affect the password if you are through with the Oracle10g installation.
Exception : : 6: cannot open shared object file: No such file or directory occurred. : : 6: cannot open shared object file: No such file or directory at Native Method at :1560 at
You could get this error message on RHEL3 x8664, RHEL4 x8664, and also on other systems. Even though you in all probability have
installed on your whole body, this error messages is complaining which it cant obtain the
On my RHEL3 x8664 system for 10g 10.1.0.3 I had to put in the i386
In order in order to meet dependencies with this i386 package, I had to put in a few other i386 RPMs too:
On my RHEL4 x8664 U1 system for 10g 10.1.0.3 I had to fit the i386
In order to meet dependencies just for this i386 package, I had to fit a few other i386 RPMs too:
On my RHEL4 x8664 U2 system for 10g R2 10.2.0.1.0 I had to setup the i386
In order to meet dependencies with this i386 package, I had to fit a few other i386 RPMs at the same time:
On my RHEL4 x86 U2 system for 10g R2 10.2.0.1.0 I had to install these RPM:
If you have this error message on Fedora Core 3 x86, install the
After I installed these RPMs I had to restart set up .. If you know a far more elegant solution to continue, please drop me an e-mail.
make - f relinksharedobj SHAREDOBJlibnmemso make1: Entering directory/u01/app/oracle/OraHome1/sysman/lib gcc - o /usr/bin/ld: crti.o: No such file: No such file or directory collect2: ld returned 1 exit status make1: Error 1
You might get this error message or even a similar one when installing 64-bit 10g on RHEL4 x8664.
On my RHEL4 U1 x8664 system for 10g 10.1.0.3 I installed the subsequent i386 RPM to fix this matter:
On my RHEL4 U2 x86-64 system for 10g R2 10.2.0.1.0 I installed these i386 RPM to fix this challenge:
You can get this error message or maybe a similar one when installing 64-bit 10g on RHEL4 x8664.
On my RHEL4 U1 x8664 system for 10g 10.1.0.3 I installed this i386 RPM to fix this issue:
On my RHEL4 U2 x8664 system for 10g R2 10.2.0.1.0 I installed the subsequent i386 RPM to fix this matter:
I saw this error when I installed 10g R2 on Fedora Core 4. I fixed it by patchting/upgrading the SELinux policy:
Error in invoking target allnoorcl of makefile. See for details.
INFO: gcc: INFO: 5: No such file or directory INFO: INFO:/u01/app/oracle/oracle/product/10.2.0/db1/bin/genorasdksh: Failed to link 10.2 INFO: make: liborasdkbase Error 1
On my RHEL4 U2 x8664 system for 10g R2 10.2.0.1.0 I installed these x8664 RPM to fix this matter:
RPM installed on the systems however you need the x8664 RPM to fix this issue. To verify which
RPM you've installed on your body, run:
Error in invoking target allnoorcl ihsodbc of makefile. See for details.
INFO: Creating 10.2 INFO: gcc: INFO: 5: No such file or directory INFO: INFO:/u01/app/oracle/oracle/product/10.2.0/db1/bin/genorasdksh: Failed to link 10.2
On my RHEL 4 U2 x86 system for 10g R2 10.2.0.1.0 I installed the subsequent RPM to fix this issue:
RPM. For i386 gleam 296 version with the
RPM. For i386 gleam 296 version in the
After that hit Retry inside the error dialog window.
Error in invoking target allnoorcl of makefile. See for details.
INFO:/usr/bin/ld:/u01/app/oracle/oracle/product/10.2.0/db1/lib/libirc.afastmemcpy.o: relocation RX8664PC32 against memcpymemopsmethod is not used when creating a shared object; recompile with - fPIC/usr/bin/ld: final link failed: Bad value collect2: ld returned 1 exit status INFO:/u01/app/oracle/oracle/product/10.2.0/db1/bin/genorasdksh: Failed to link 10.2
This error appears when installing 10g R2 10.2.0.1.0 on RHEL4 x8664. Make sure to upgrade to RHEL4 U3 or download the
There may be many reasons because of this error. For example, this will happen during ASM instance startup when the
RPM is just not installed about the system.
127.0.0.1 localhost
The information provided with this website comes without warranty regardless of the sort and is distributed AS IS. Every effort may be made to produce the information as accurate as you can, but no warranty or fitness is implied. The information can be incomplete, might have errors or might have become expired. The use of this data described herein will be your responsibility, and to work with it a highly effective environments achieve this at your own risk.
Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012
Like this document? Why not share!
Email sent successfully!
Understanding how your body resources are widely-used and how well they perform might be extremely valuable when you plan your infrastructure, making picking a the computer a pivotal decision that can influence your IT strategy for ages to come. Throughout our CPU and RAM tests, we found the open-source Red Hat Enterprise Linux 6 solution performed too or much better than Microsoft Windows Server 2012. In our SPEC CPU2006 tests, the Red Hat Enterprise Linux 6 solution achieved consistently higher scores as opposed to Windows Server 2012 solution. When we used the LINPACK benchmark to try floating point performance of CPUs, we also learned that tuning the os allowed us for getting even greater performance out from the Red Hat Enterprise Linux 6 system. In our memory bandwidth tests, the Red Hat Enterprise Linux 6 solution outperformed the Windows Server 2012 solution at mid-range thread counts.
By choosing an os that can deliver strong performance on all subsystems out on the box and increase performance more when tuned, you are able to ensure that that you are giving your applications hidden resources to execute well and providing your company with a compelling foundation for future growth.
1. APRIL 2013 A PRINCIPLED TECHNOLOGIES TEST REPORT Commissioned by Red Hat, Inc. COMPARING CPU AND MEMORY PERFORMANCE: RED HAT ENTERPRISE LINUX 6 VS. MICROSOFT WINDOWS SERVER 2012 An operating systems power to effectively manage and make use of server hardware often defines system and application performance. Processors with multiple cores and random access memory RAM represent both most vital subsystems that could affect the performance of business applications. Selecting the best performing operating-system can help your hardware achieve its maximum potential and enable your critical applications running faster. To help you make that critical decision, Principled Technologies compared the CPU and RAM performance of Red Hat Enterprise Linux 6 and Microsoft Windows Server 2012 using three benchmarks: SPEC CPU2006, LINPACK, and STREAM. We learned that Red Hat Enterprise Linux 6 delivered better CPU and RAM performance in virtually every test, outperforming its Microsoft competitor within out-of - box and optimized configurations.
2. A Principled Technologies test report 2Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 BETTER CPU AND RAM PERFORMANCE We compared CPU and RAM performance on two os: Red Hat Enterprise Linux 6 and Microsoft Windows Server 2012. For our comparison, we used the SPEC CPU2006 benchmark and LINPACK benchmark to evaluate the CPU performance from the solutions while using the different os, plus the STREAM benchmark to try the memory bandwidth in the two solutions. For each test, we first configured both solutions with out-of-box default settings, therefore we tested those solutions using multiple tuning parameters to offer optimized results. We ran each test thrice and report the results from your median run. For detailed system configuration information, see Appendix A. For particularly how we tested, see Appendix B. Testing the CPU with SPEC CPU2006 and LINPACK When we tested CPU performance, we found that this server running Red Hat Enterprise Linux 6 scored as much as 5.one percent higher compared to the Microsoft Windows Server 2012 solution for the SPEC CPU2006 benchmark, both out-of - box and optimized. The SPEC CPU2006 benchmark contains two benchmark suites, as both versions focuses on the different component of compute-intensive performance. SPEC CINT2006 measures and compares compute-intensive integer performance, while SPEC CFP2006 measures and compares compute-intensive floating-point performance. A rate version of each and every, which runs multiple instances on the benchmark to evaluate server performance, is additionally available. We ran the speed version of such benchmarks. Figures 1 and a couple of show the scores the systems achieved on both parts with the benchmark. For detailed test results, see Appendix C. Figure 1: SPEC CPU2006 results, in SPEC CINT2006 scores, with the two solutions. Higher numbers are better. 639 640 623 621618 619 593 591 0 100 200 300 400 500 600 700 Out-of-box Optimized Out-of-box Optimized SPECintrate2006 SPECintratebase2006 SPECCINT2006score SPEC CINT2006 scores Red Hat Enterprise Linux 6 Microsoft Windows Server 2012
3. A Principled Technologies test report 3Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 Figure 2: SPEC CPU2006 results, in SPEC CFP2006 scores, for your two solutions. Higher numbers are better. 422 422 408 404406 405 399 397 0 50 100 150 200 250 300 350 400 450 Out-of-box Optimized Out-of-box Optimized SPECfprate2006 SPECfpratebase2006 SPECCFP2006score SPEC CFP2006 scores Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Measuring floating point computing power assesses the performance of high- performance systems that requirement to do heavy work. We used the LINPACK benchmark to evaluate the floating point performance with the platforms out-of - box and optimized, which shows the way a system performs under a high load. As Figure 3 shows, Red Hat Enterprise Linux 6 outperformed Windows Server 2012 when utilizing all 16 physical cores on our test system. In addition, tuning the Red Hat Enterprise Linux 6 system increased performance steadily when moving from 4 threads to 16 threads, while optimizing Windows Server 2012 had little affect on its performance at 16 threads. In our tests, we enabled Intel Hyper-Threading Technology, but bound the workload to at many one thread per physical core. We discovered that allocating every available thread beyond 32 on the test would not increase LINPACK performance for either platform. For detailed test results, see Appendix C.
4. A Principled Technologies test report 4Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 Figure 3: LINPACK floating point performance results for your two os solutions, both out-of-the-box and optimized. Higher numbers are better. 0 50 100 150 200 250 300 1 2 4 8 16 PerformanceGFlops Number of threads LINPACK: Average floating point performance Red Hat Enterprise Linux 6 optimized Microsoft Windows Server 2012 optimized Red Hat Enterprise Linux 6 out-of-box Microsoft Windows Server 2012 out-of- box Testing RAM performance Because os manage physical memory, the effective memory bandwidth, which can be the rate from which a system can understand or write data from memory, is usually dependent on OS capabilities and configuration. We used the STREAM benchmark to look at the memory bandwidth both Red Hat Enterprise Linux 6 and Microsoft Windows Server 2012 delivered. As Figures 4 and 5 show, both solutions performed comparably in this region, both out-of-box and optimized. In our tests, we enabled Intel Hyper-Threading Technology, but bound the workload to at many one thread per physical core. We discovered that allocating every available thread from 32 towards the test failed to increase STREAM performance for either platform. For detailed test results, see Appendix C.
5. A Principled Technologies test report 5Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 Figure 4: Out-of-box memory bandwidth comparison utilizing the STREAM benchmark. 0 10, 000 20, 000 30, 000 40, 000 50, 000 60, 000 70, 000 80, 000 Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad 1 2 4 8 16 SustainedmemorybandwidthMB/s STREAM tests and volume of cores Memory bandwidth comparison: Out-of-box configurations Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Figure 5: Optimized memory bandwidth comparison utilizing the STREAM benchmark. 0 10, 000 20, 000 30, 000 40, 000 50, 000 60, 000 70, 000 80, 000 Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad Copy Scale Add Triad 1 2 4 8 16 SustainedmemorybandwidthMB/s STREAM tests and quantity of cores Memory bandwidth comparison: Optimized configurations Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 WHAT WE TESTED About Red Hat Enterprise Linux 6 Red Hat Inc. positions Red Hat Enterprise Linux 6 like a prime competitor to proprietary systems found in enterprise data centers. Red Hat Enterprise Linux 6 is designed to offer performance and scalability for both big and small servers with documented scalability nearly 4, 096 CPUs and 64 terabytes of RAM. It
6. A Principled Technologies test report 6Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 provides native support for that majority of the most up-to-date and most important enterprise data center technologies, like 40Gb Ethernet networking and KVM virtualization at the same time as InfiniBand, FCoE, and iSCSI protocols. According to Red Hat, Red Hat as well as hardware partners are enabling reliability, availability, serviceability RAS, and scalability features to aid minimize downtime, increase availability, and protect data. Red Hat includes included in its Linux offering tested, open source applications. For details about Red Hat Enterprise Linux 6, see About SPEC CPU2006 SPEC CPU2006 is definitely an industry-standard benchmark using a CPU-intensive workload to worry a systems processors, memory subsystem, and compiler. SPEC CPU2006 encompasses home equity loans tests: SPEC CINT2006, which reports results in the SPECintrate2006 and SPECintratebase2006 scores, and SPEC CFP2006, which reports results in the SPECfprate2006 and SPECfpratebase2006 scores. These scores help compare an array of hardware. For details about the SPEC CPU2006 benchmark, visit /cpu2006/. About LINPACK The LINPACK benchmark runs a course that solves a method of linear equations to look at the floating point rate of execution of something. Often accustomed to test the performance of supercomputers, LINPACK will help determine the peak performance that something is able to by using complex calculations to push the processor. For more details about LINPACK, visit /project/linpack/. About STREAM STREAM is undoubtedly an industry-standard benchmark that measures the memory bandwidth of a head unit. The benchmark measures real-world sustained memory bandwidth for typical operations on data, rather then peak memory performance rates. For additional information about the STREAM benchmark, visit IN CONCLUSION Understanding how the body resources are used and how well they perform could be extremely valuable when you plan your infrastructure, making selecting the main system a pivotal decision that may influence your IT strategy for several years to come. Throughout our CPU and RAM tests, we found how the open-source Red Hat Enterprise Linux 6 solution performed too or much better than Microsoft Windows Server 2012. In our SPEC CPU2006 tests, the Red Hat Enterprise Linux 6 solution achieved consistently higher scores compared to Windows Server 2012 solution. When we
7. A Principled Technologies test report 7Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 used the LINPACK benchmark to find out floating point performance of CPUs, we also found out that tuning the computer allowed us to have even greater performance out in the Red Hat Enterprise Linux 6 system. In our memory bandwidth tests, the Red Hat Enterprise Linux 6 solution outperformed the Windows Server 2012 solution at mid- range thread counts. By choosing an main system that can deliver strong performance on all subsystems out on the box and increase performance a lot more when tuned, you may ensure that you're giving your applications the required resources to do well and providing your enterprise with an excellent foundation for future growth.
8. A Principled Technologies test report 8Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 APPENDIX A SYSTEM CONFIGURATION INFORMATION Figure 6 shows it configuration information to the server we utilized in our tests. System Dell PowerEdge R720 Power supplies Total number 2 Vendor and model number Dell D750E-S1 Wattage of each and every W 750 Cooling fans Total number 6 Vendor and model number AVC DBTC0638B2V Dimensions h x w of each and every 2.5 x 2.5 Volts 12 Amps 1.2 General Number of processor packages 2 Number of cores per processor 8 Number of hardware threads per core 2 System power management policy Performance Per Watt DAPC or Performance Per Watt OS see text CPU Vendor Intel Name Xeon Model number E5-2690 Stepping 6 Socket type 2011LGA Core frequency GHz 2.90 Bus frequency 100 L1 cache 32 KB I 32 KB D per core L2 cache 256 KB on chip per core L3 cache 20 MB Platform Vendor and model number Dell PowerEdge R720 Motherboard model number OM1GCR BIOS name and version Dell 1.5.1 BIOS settings Default Memory modules Total RAM in system GB 128 Vendor and model number Samsung M393B1K70BH1-CH9 Type PC3-10600 Speed MHz 1, 333 Speed running within the system MHz 1, 333 Timing/Latency tCL-tRCD-tRP-tRASmin 9-9-9-36 Size GB 8
9. A Principled Technologies test report 9Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 System Dell PowerEdge R720 Number of RAM modules 8 Chip organization Double-sided Rank Dual Operating system 1 Name Red Hat Enterprise Linux 6.4 Filesystem Ext4 Kernel 6.x8664, or 6.x8664 see text Language English Operating system 2 Name Windows Server 2012 Datacenter Edition Build number 9200 Filesystem NTFS Kernel ACPI x64-based PC Language English Graphics Vendor and model number Matrox G200eR Graphics memory MB 16 Driver Matrox Graphics, Inc. 2.3.3.0 8/19/2011 RAID controller Vendor and model number Dell PERC H710P Mini Firmware version 21.1.0-007 Cache size 1 GB RAID configuration OS 1: RAID 1 configuration of two type-1 disks. OS 2: RAID 1 configuration of two type-1 disks. Swap/Utility: Raid 1 configuration of twotype-2 disks. Hard drives type 1 Vendor and model number Seagate Savio ST973451SS Number of drives 4 Size GB 73 Buffer size MB 16 RPM 15K Type SAS Hard drives type 2 Vendor and model number Dell WD300BKHG-18A29V0 Number of drives 2 Size GB 300 Buffer size MB 16 RPM 10K Type SAS
10. A Principled Technologies test report 10Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 System Dell PowerEdge R720 Ethernet adapters 1 Vendor and model number Intel Gigabit 4P I350-t rNDC Type Internal Ethernet adapters 2 Vendor and model number Intel Ethernet Server Adapter X520-SR1 Type PCIe Optical drives Vendor and model number TEAC DV-28SW Type DVD-ROM USB ports Number 4 external, 1 internal Type 2.0 Figure 6: Configuration information for that test system.
11. A Principled Technologies test report 11Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 APPENDIX B HOW WE TESTED In this section, we present the step-by-steps for the way you tested. First, we cover main system installation and optimization settings for optimized solution testing just for both solutions. Then, we detail how to setup and test using each benchmark. Installing Red Hat Enterprise Linux 6.4 We installed Red Hat Enterprise Linux 6.4 around the first of two logical volumes, configured with two mirrored disks. 1. Insert and boot on the Red Hat Enterprise Linux 6.4 x8664 installation DVD. 2. At the welcome screen, select Install or upgrade a pre-existing system, and press Enter. 3. At the Media test screen, select Skip, and press Enter. 4. At the Red Hat Enterprise Linux 6 title screen, click Next. 5. At the Choose an Installation Language screen, select English, and then click Next. 6. At the Keyboard Type screen, select English, and then click Next. 7. At the Storage Devices screen, select Basic Storage Devices, and then click Next. 8. If a warning for device initialization appears, select Yes, discard any data. 9. At the Name the Computer screen, type the host name, and click on Configure Network. 10. At the Network Connections screen, choose the servers main or management network interface, and then click Edit. 11. At the Editing network interface screen, check Connect Automatically. 12. On the same screen, pick the IPv4 Settings tab, modify the Method to Manual, and click on Add. 13. On the same screen, go into the IP address, Netmask, Gateway, and DNS server. Click Apply. 14. Click Close for the Network Connections screen, and click on Next around the Name the Computer screen. 15. At the Time zone selection screen, pick the appropriate time zone, and then click Next. 16. Enter the main password from the Root Password and Confirm fields, and click on Next. 17. At the Assign Storage Devices screen, from your list inside left column, choose the Linux disk, and then click the arrow copying the device towards the right column. Next on the Linux disk, select the Boot radio button, and click on Next. 18. At the Partition selection screen, select Replace Existing Linux Systems, and click on Next. 19. If a warning appears, click Write changes to disk. 20. At the default installation screen, click Next to begin set up .. 21. At the Congratulations screen, click Reboot. 22. After it reboots, signing in as root. Installing Microsoft Windows Server 2012 Datacenter We installed Microsoft Windows Server 2012 Datacenter for the second of two logical volumes, configured with two mirrored disks. 1. Insert and boot through the Windows Server 2012 Datacenter installation DVD. 2. At the first Window Setup screen, maintain the defaults for installation language, time/currency format, and keyboard input method. Click Next. 3. At the second Windows Setup screen, click Install now. 4. At the third Windows Setup screen, type in the Windows activation key, and click on Next.
12. A Principled Technologies test report 12Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 5. At the fourth Windows Setup screen, find the Windows Server 2012 Datacenter Server that has a GUI, and click on Next. 6. At the fifth Windows Setup screen, simply select the checkbox to simply accept the license term, and then click Next. 7. At the sixth Windows Setup screen, click Custom: Install Windows only advanced. 8. At the seventh Windows Setup screen, select Drive 2 because the Windows installation drive, and then click Next to start installation. 9. The system will reboot. At the Settings screen, type in the password for that Administrator twice, and click on Finish. Optimizing os configurations Before running tests for that optimized Red Hat Enterprise Linux 6 configuration, run the subsequent two bash scripts. Red Hat Enterprise Linux 6.4 uses the tuned utility with all the throughput-performance profile to configure the filesystems for better performance and to own the CPUs at top rated while the CPU power profile is under OS control. Run the next command to activate this tuned profile. tuned-adm profile throughput-performance !/bin/bash For the optimized-configuration tests, disable unneeded services for i in abrt-ccpp abrt-oops abrtd acpid atd auditd autofs avahi-daemon cgconfig crond cups haldaemon irqbalance kdump libvirt-guests mcelogd mdmonitor messagebus portreserve postfix rhnsd rhsmcertd rpcbind rpcgssd rpcidmapd certmonger netfs sysstat; do service i stop done service lvm2-monitor force-stop end of Before running tests for your optimized Windows Server 2012 configuration, run these batch script. In particular, the OS power profile is defined to High performance along with the desktop GUI is configured for good performance. rem Note: script closely adapted from rem /2010/09/13/creating-a-windows-7-template-for-vmware- view/rem Version dated 20120205 reg load hkutemp reg ADD hkutempSoftwarePoliciesMicrosoftWindowsControl PanelDesktop/v/d /f reg ADD hkutempSoftwarePoliciesMicrosoftWindowsControl PanelDesktop/v ScreenSaveTimeOut/d 600/f reg ADD hkutempSoftwarePoliciesMicrosoftWindowsControl PanelDesktop/v ScreenSaverIsSecure/d 1/f reg ADD hkutempSoftwareMicrosoftWindowsCurrentVersionPoliciesSystem/v Wallpaper/d /f reg ADD hkutempSoftwareMicrosoftWindowsCurrentVersionInternet SettingsCache/v Persistent/t REGDWORD/d 0x0/f
13. A Principled Technologies test report 13Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 reg ADD hkutempSoftwareMicrosoftFeeds/v SyncStatus/t REGDWORD/d 0x0/f reg ADD hkutempSoftwareMicrosoftWIndowsCurrentVersionPoliciesExplorer/v HideSCAHealth/t REGDWORD/d 0x1/f reg unload hkutemp reg ADD HKEYLOCALMACHINESOFTWAREPoliciesMicrosoftInternet ExplorerMain/v DisableFirstRunCustomize/t REGDWORD/d 0x1/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetControlSession ManagerMemory ManagementPrefetchParameters/v EnableSuperfetch/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESOFTWAREPoliciesMicrosoftWindowsWindowsUpdateAU/v NoAutoUpdate/t REGDWORD/d 0x1/f reg ADD HKEYLOCALMACHINESOFTWAREPoliciesMicrosoftWindows NTSystemRestore/v DisableSR/t REGDWORD/d 0x1/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetservicesDisk/v TimeOutValue/t REGDWORD/d 200/f reg ADD HKEYLOCALMACHINESOFTWAREImage/v Revision/t REGSZ/d 1.0/f reg ADD HKEYLOCALMACHINESOFTWAREImage/v Virtual/t REGSZ/d Yes/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogApplication/v MaxSize/t REGDWORD/d 0x100000/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogApplication/v Retention/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESystemCurrentControlSetControlNetworkNewNetworkWindo wOff/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogSystem/v MaxSize/t REGDWORD/d 0x100000/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogSystem/v Retention/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogSecurity/v MaxSize/t REGDWORD/d 0x100000/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetserviceseventlogSecurity/v Retention/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetControlCrashControl/v CrashDumpEnabled/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESOFTWAREMicrosoftWindowsCurrentVersionpoliciesExplo rer/v NoRecycleFiles/t REGDWORD/d 0x1/f reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetControlTerminal Server/v fDenyTSConnections/t REGDWORD/d 0x0/f
14. A Principled Technologies test report 14Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 reg ADD HKEYLOCALMACHINESYSTEMCurrentControlSetControlTerminal ServerWinStationsRDP-Tcp/v UserAuthentication/t REGDWORD/d 0x0/f reg ADD HKEYLOCALMACHINESoftwareMicrosoftWindowsCurrentVersionpoliciessyste m/v EnableLUA/t REGDWORD/d 0x0/f reg Add HKEYLOCALMACHINESoftwarePoliciesMicrosoftWindowsSideshow/v Disabled/t REGDWORD/d 0x1/f Powershell Set-Service BDESVC -startuptype disabled Powershell Set-Service wbengine -startuptype disabled Powershell Set-Service DPS -startuptype disabled Powershell Set-Service UxSms -startuptype disabled Powershell Set-Service Defragsvc -startuptype disabled Powershell Set-Service HomeGroupListener -startuptype disabled Powershell Set-Service HomeGroupProvider -startuptype disabled Powershell Set-Service iphlpsvc -startuptype disabled Powershell Set-Service MSiSCSI -startuptype disabled Powershell Set-Service swprv -startuptype disabled Powershell Set-Service CscService -startuptype disabled Powershell Set-Service SstpSvc -startuptype disabled Powershell Set-Service wscsvc -startuptype disabled Powershell Set-Service SSDPSRV -startuptype disabled Powershell Set-Service SysMain -startuptype disabled Powershell Set-Service TabletInputService -startuptype disabled Powershell Set-Service Themes -startuptype disabled Powershell Set-Service upnphost -startuptype disabled Powershell Set-Service VSS -startuptype disabled Powershell Set-Service SDRSVC -startuptype disabled Powershell Set-Service WinDefend -startuptype disabled Powershell Set-Service WerSvc -startuptype disabled Powershell Set-Service MpsSvc -startuptype disabled Powershell Set-Service ehRecvr -startuptype disabled Powershell Set-Service ehSched -startuptype disabled Powershell Set-Service WSearch -startuptype disabled Powershell Set-Service wuauserv -startuptype disabled Powershell Set-Service Wlansvc -startuptype disabled Powershell Set-Service WwanSvc -startuptype disabled bcdedit/set BOOTUX disabled vssadmin delete shadows/All/Quiet Powershell disable-computerrestore - drive c: netsh advfirewall set allprofiles state off powercfg - H OFF powercfg - setactive 8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c net stop sysmain fsutil behavior set DisableLastAccess 1 schtasks/change/TN MicrosoftWindowsDefragScheduledDefrag/Disable schtasks/change/TN MicrosoftWindowsSystemRestoreSR/Disable schtasks/change/TN MicrosoftWindowsRegistryRegIdleBackup/Disable schtasks/change/TN MicrosoftWindows DefenderMPIdleTask/Disable schtasks/change/TN MicrosoftWindows DefenderMP Scheduled Scan/Disable schtasks/change/TN MicrosoftWindowsMaintenanceWinSAT/Disable
15. A Principled Technologies test report 15Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 rem End of TESTING WITH SPEC CPU2006 Installing the SPEC CPU2006 software around the Red Hat Enterprise Linux 6.4 computer We used version 1.2 with the SPEC CPU2006 distribution with binaries, run-time libraries and run-time options taken in the Intel Linux 64-bix distribution, which uses the Intel ic13 compiler. Specifically, we used configuration with the x64 AVX Intel processor-type for your rate and speed SPEC CPU runs. 1. Log onto Red Hat Enterprise Linux 6.4. 2. Create a scratch partition for SPEC CPUs intermediate results with an unused volume here/dev/sdb: parted/dev/sdb mklabel gpt parted/dev/sdb mkpart primary 1 - 1 parted/dev/sdb name 1 scratch 4/dev/sdb1 mkdir/scratch 3. Configure the computer to mount the scratch partition at directory/scratch. Edit the file/etc/fstab and add this lines towards the end. Scratch space for SPEC CPU/dev/sdb1/scratch ext4 defaults 0 0 4. Mount the scratch partition. mount/scratch 5. Install the GCC compiler also as the run-time libraries for 32-bit programs by adding this packages: gcc, libc.i686, libgcc.i686, libstdc.i686, and glibc-devel.i686. 6. Create a local SPEC CPU directory. mkdir/opt/cpu2006 7. Mount the SPEC CPU2006 v1.2 installation CD or ISO image to/mnt 8. Run the SPEC CPU2006 installation script on the installation directory. Answer Y to both questions. cd/mnt sh/opt/cpu2006 9. Extract the Intel 18feb2013-linux64 binaries, libraries and scripts in the archive into/opt/cpu2006. cd/opt/cpu2006 tar Jxf 10. Modify file permissions for your SPEC CPU executables and shared libraries. cd/opt/cpu2006 find benchspec/CPU2006 - name 64.27dec2012 xargs chmod 755 find sh libs - name xargs chmod 755 11. Modify the run-time configuration files to ensure SPEC CPU uses the scratch directory for intermediate results. Go to directory/opt/cpu2006/config and add the next line to both and following your line reportable1. outputroot/scratch
16. A Principled Technologies test report 16Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 12. Copy the contents with the SPEC CPU executables directory to/scratch. rm r/scratch/benchspec/cp pr/opt/cpu2006/benchspec/scratch/benchspec Running the SPEC CPU2006 tests about the Red Hat Enterprise Linux 6.4 main system We used Intels batch script, running the SPEC CPU2006 FP and INT rate and speed tests. ulimit - s unlimited acat/proc/cpuinfo grep processor wc - l rm - rf specperl bcat cexpr a2 echo echo Running rate that has a copies with an SSE4.2 system that has a topology of b echo if NUMA 0 then if -e/sys/kernel/mm/redhattransparenthugepage/enabled then echo always /sys/kernel/mm/redhattransparenthugepage/enabled runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define no-numa - -define THPenabled - o all int runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define no-numa - -define THPenabled - o all fp else runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define no-numa - o all int runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define no-numa - o all fp fi else echo 1 /proc/sys/vm/dropcaches if -e/sys/kernel/mm/redhattransparenthugepage/enabled then echo always /sys/kernel/mm/redhattransparenthugepage/enabled numactl - -interleaveall runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define invokewithinterleave - -define dropcaches - - define THPenabled - o all int numactl - -interleaveall runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define invokewithinterleave - -define dropcaches - - define THPenabled - o all fp else
17. A Principled Technologies test report 17Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 numactl - -interleaveall runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define invokewithinterleave - -define dropcaches - o all int numactl - -interleaveall runspec - -define external - -rate a - c 27dec2012 - -define smt - -define coresc - -define b - -define invokewithinterleave - -define dropcaches - o all fp fi fi Installing the SPEC CPU2006 software around the Windows Server 2012 operating-system We used version 1.2 from the SPEC CPU2006 distribution with binaries, run-time libraries and run-time options taken through the Intel Windows 64-bit distribution, which uses the Intel ic13 compiler. Specifically, we used configuration with the x64 AVX Intel processor-type for your rate and speed SPEC CPU runs. 1. Log in as administrator to develop a scratch partition for SPEC CPUs intermediate results. 2. Open the Server Manager. 3. Select File and Storage Services. 4. Select Disks. 5. From the report on disks, right-click with your Scratch device, select Reset Disk, and then click Yes to erase the information. 6. From the number of disks, right-click on the Scratch device, and select New Volume 7. On the Server and Disk screen, find the disk, and then click Next. 8. Click OK to initialize the disk which has a GPT label. 9. On the Size screen, keep your defaults, and click on Next. 10. On the Drive Letter or Folder screen, Select The following folder, and enter c:scratch. Click OK to produce this folder. 11. On the File System Settings screen, select NTFS. 12. On the same screen, enter a Volume label of Scratch, and then click Next. 13. On the Confirmation screen, click Create. 14. Close the Server Manager. 15. Restart the server. 16. Create a local SPEC CPU directory. mkdir c:cpu2006 17. Mount the SPEC CPU2006 v1.2 installation CD or ISO image as drive d: 18. Run the SPEC CPU2006 installation script on the installation directory. Answer Y to both questions. cd d: c:cpu2006 19. Extract the Intel sep2102-winx64 binaries, libraries and scripts from your archive into c:cpu2006. 20. Modify the run-time configuration files in order that SPEC CPU uses the scratch directory for intermediate results. Go to directory c:cpu2006config and add the next line to both and
18. A Principled Technologies test report 18Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 as soon as the line reportable1. outputroot/scratch 21. Copy the items in directory c:cpu2006benchspec to d:scratchbenchspec. Running the SPEC CPU2006 tests within the Windows Server 2012 operating-system We used Intels batch script, to operate the SPEC CPU2006 FP and INT rate and speed tests. call set/a numcores%NUMBEROFPROCESSORS%/2 set OMPNUMTHREADS%numcores% set KMPAFFINITYgranularityfine, scatter call runspec - c - T all - o asc int call runspec - c - T all - o asc fp set OMPNUMTHREADS set KMPAFFINITY call runspec - -rate %NUMBEROFPROCESSORS% - -define smt-on - -define cores%numcores% - c - T all - o asc int call runspec - -rate %NUMBEROFPROCESSORS% - -define smt-on - -define cores%numcores% - c - T all - o asc fp call runspec - -loose - n 1 - -rate 1 - -define smt-on - -define cores%numcores% - c - T base - o asc int call runspec - -loose - n 1 - -rate 1 - -define smt-on - -define cores%numcores% - c - T base - o asc fp TESTING WITH LINPACK Installing LINPACK benchmark about the Red Hat Enterprise Linux 6.4 os We used Intels binary, version 11.0.3 for Linux x64, for that LINPACK benchmark. 1. Log onto Red Hat Enterprise Linux 6.4. 2. Uncompress the archive. tar zxf 3. The benchmark code is within directory linpack11.0.3/benchmarks/linpack. cd linpack11.0.3/benchmarks/linpack Running the LINPACK benchmark around the Red Hat Enterprise Linux 6.4 main system We adapted Intels sample batch script and used its unmodified input data to perform the LINPACK benchmark. !/bin/bash Adapted from Intels sample run script Run parameters are read from lininputxeon64 Setting up affinity for better threading performance
19. A Principled Technologies test report 19Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 MYKMPAFFINITYnowarnings, granularityfine, verbose, explicit, proclist p1 0 p2 0, 1 p4 0, 2, 1, 3 p8 0, 2, 4, 6, 1, 3, 5, 7 p16 0, 2, 4, 6, 8, 10, 12, 14, 1, 3, 5, 7, 9, 11, 13, 15 p32 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 2 6, 27, 28, 29, 30, 31 archxeon64 for i in 1 2 4 8 16 32; do echo 3 /proc/sys/vm/dropcaches echo Threads i export OMPNUMTHREADSi export KMPAFFINITYMYKMPAFFINITYpi echo Affinity KMPAFFINITY date./xlinpackarch lininputarch echo - n Done: date done tee - a end of shell script lininputxeon64 Sample IntelR Optimized LINPACK Benchmark data file lininputxeon64 IntelR Optimized LINPACK Benchmark data 15 volume of tests 1000 2000 5000 10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000 problem sizes 1000 2000 5008 10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000 leading dimensions 4 2 2 2 2 2 2 2 2 2 1 1 1 1 1 times to operate a test 4 4 4 4 4 4 4 4 4 4 4 1 1 1 1 alignment values in KBytes Installing the LINPACK benchmark around the Windows Server 2012 computer To run the LINPACK benchmark, we adapted Intels sample batch script and modified the Windows input data to fit that used within the Red Hat Linux tests. 1. Log onto Windows Server 2012. 2. Uncompress the archive. unzip 3. The benchmark code was in directory linpack11.0.3/benchmarks/linpack. cd wlpkp11.0.3.008linpack11.0.3benchmarkslinpack Running the LINPACK benchmark around the Windows Server 2012 os We used this batch script and input data to perform the LINPACK benchmark. echo off:: echo Adapted from Intels sample run script
20. A Principled Technologies test report 20Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 SETLOCAL EnableDelayedExpansion rem Argument 1 would be the tag with the runs output file set rem Setting up BASE affinity for better threading performance set MYKMPAFFINITYnowarnings, granularityfine, verbose, explicit, proclist rem Setting route to OpenMP library set 64compiler;%PATH% rem My processor affinities set p10 set p20, 16 set p40, 2, 16, 18 set p80, 2, 4, 6, 16, 18, 20, 22 set p160, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30 set p320, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 2 7, 28, 29, 30, 31 date/t time/t echo Running Output could be seen in %fff%. for %%i in 1 2 4 8 16 32 do echo %%i:: echo!p%%i! set OMPNUMTHREADS%%i set KMPAFFINITY%MYKMPAFFINITY%!p%%i! echo!KMPAFFINITY! new-lininputxeon64 %fff% date/t %fff% time/t %fff% echo Done: date/t time/t ENDLOCAL echo on rem End of batch script new-lininputxeon64 Modified IntelR Optimized LINPACK Benchmark data file lininputxeon64 IntelR Optimized LINPACK Benchmark data 15 amount of tests 1000 2000 5000 10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000 problem sizes 1000 2000 5008 10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000 leading dimensions 4 2 2 2 2 2 2 2 2 2 1 1 1 1 1 times to own a test 4 4 4 4 4 4 4 4 4 4 4 1 1 1 1 alignment values in KBytes
21. A Principled Technologies test report 21Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 TESTING WITH STREAM Installing the Stream software about the Red Hat Enterprise Linux 6.4 operating-system We used version 5.10 in the Stream distribution from /and compiled it while using Intel ic13 compiler for Linux. 1. Log onto Red Hat Enterprise Linux 6.4. 2. Uncompress the original source-code archive into your directory stream-5.10. 3. Create sf, the foreground version of Stream 10 iterations, with all the Intel C compiler: icl O3 xAVX DNTIMESS10 stream.c o sf 4. Create sb, the foreground version of Stream 100 iterations, with all the Intel C compiler: 5. icl O3 xAVX DNTIMESS100 stream.c o sb Running the Stream tests within the Red Hat Enterprise Linux 6.4 os In order to complete the multi-core tests, we followed the guidance through the Stream documentation, and used multiple copies We used Intels batch script to operate the Stream FP and INT rate and speed tests. !/bin/bash exec 2 /dev/null date pkill - 9 sf /dev/null 2 1 pkill - 9 sb /dev/null 2 1 sleep 3 for size in 1 2 4 8 16 32; do echo 3 /proc/sys/vm/dropcaches printf nn printf Doing sizen printf Launching if size - ne 1 ; then for i in seq size-1 ; do printf ii % 2 numactl - -membind i % 2 - -physcpubind i nice - n - 10./sb /dev/null 2 1 done fi sleep 3 printf 00n numactl - -membind 0 - -physcpubind 0 nice - n - 10./sf pkill - 9 sf /dev/null 2 1 pkill - 9 sb /dev/null 2 1 wait
22. A Principled Technologies test report 22Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 done printf nn date exit 0 end of shell script Installing the Stream software for the Windows Server 2012 main system We used version 5.10 on the Stream distribution from /stream/FTP/Code/and compiled it with all the Intel ic13 compiler for Windows within the Microsoft Studio environment. We also installed the psutil tools from We replaced the Unix/Linux-style timer with one right for Windows begin to see the patch-file, below, for details. 1. Create a local Stream directory. mkdir stream-5.10 2. Copy the Stream 5.10 source files into this directory. 3. We modified the cause code for stream.c, version 5.10, to utilize a Windows-style routine to measure time. The patch file, stream.c-diff, with the modifications follows. You can apply the modification with all the Unix/Linux command: patch stream.c-diff stream.c-diff 44c44 //include unistd.h - include unistd.h 48, 73c48 //include sys/time.h include windows.h typedef int ssizet; //////////////////double PCFreq 0.0; int64 CounterStart 0; void StartCounter LARGEINTEGER li; if!QueryPerformanceFrequency li printfQueryPerformanceFrequency failed!n; PCFreq ; QueryPerformanceCounter li; CounterStart ; double mysecond
23. A Principled Technologies test report 23Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 LARGEINTEGER li; QueryPerformanceCounter li; return -CounterStart/PCFreq; ///////////////////- include sys/time.h 133c108 define NTIMES 100 - - define NTIMES 10 242d216 StartCounter; 441, 442d414 if 0 457d428 endif 4. Create, the foreground version of Stream 10 iterations, using the Intel C compiler. icl/O3/QxAVX DNTIMES10 stream.c o 5. Create, the foreground version of Stream 100 iterations, while using Intel C compiler. icl/O3/QxAVX DNTIMES100 stream.c o Running the Stream tests within the Windows Server 2012 computer We used Intels batch script to operate the Stream FP and INT rate and speed tests. echo off set echo %fff% del %fff% NUL 2 1 rem rem 1 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 1 %fff% echo 1 call fg-stream %fff% rem rem 2 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 2 %fff%
24. A Principled Technologies test report 24Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 echo 2 start/high/node 1/affinity 0x0001/b NUL choice/t 20/c yn/d y NUL 2 1 call fg-stream %fff% rem rem 4 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 4 %fff% echo 4 start/high/node 1/affinity 0x0001/b NUL start/high/node 0/affinity 0x0002/b NUL start/high/node 1/affinity 0x0002/b NUL choice/t 20/c yn/d y NUL 2 1 call fg-stream %fff% rem rem 8 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 8 %fff% echo 8 start/high/node 1/affinity 0x0001/b NUL start/high/node 0/affinity 0x0002/b NUL start/high/node 1/affinity 0x0002/b NUL start/high/node 0/affinity 0x0004/b NUL start/high/node 1/affinity 0x0004/b NUL start/high/node 0/affinity 0x0008/b NUL start/high/node 1/affinity 0x0008/b NUL choice/t 20/c yn/d y NUL 2 1 call fg-stream %fff% rem rem 16 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 16 %fff% echo 16 start/high/node 1/affinity 0x0001/b NUL start/high/node 0/affinity 0x0002/b NUL start/high/node 1/affinity 0x0002/b NUL
25. A Principled Technologies test report 25Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 start/high/node 0/affinity 0x0004/b NUL start/high/node 1/affinity 0x0004/b NUL start/high/node 0/affinity 0x0008/b NUL start/high/node 1/affinity 0x0008/b NUL start/high/node 0/affinity 0x0010/b NUL start/high/node 1/affinity 0x0010/b NUL start/high/node 0/affinity 0x0020/b NUL start/high/node 1/affinity 0x0020/b NUL start/high/node 0/affinity 0x0040/b NUL start/high/node 1/affinity 0x0040/b NUL start/high/node 0/affinity 0x0080/b NUL start/high/node 1/affinity 0x0080/b NUL choice/t 20/c yn/d y NUL 2 1 call fg-stream %fff% rem rem 32 pskill NUL 2 1 pskill NUL 2 1 choice/t 5/c yn/d y NUL 2 1 echo %fff% echo 32 %fff% echo 32 start/high/node 1/affinity 0x0001/b NUL start/high/node 0/affinity 0x0002/b NUL start/high/node 1/affinity 0x0002/b NUL start/high/node 0/affinity 0x0004/b NUL start/high/node 1/affinity 0x0004/b NUL start/high/node 0/affinity 0x0008/b NUL start/high/node 1/affinity 0x0008/b NUL start/high/node 0/affinity 0x0010/b NUL start/high/node 1/affinity 0x0010/b NUL start/high/node 0/affinity 0x0020/b NUL start/high/node 1/affinity 0x0020/b NUL start/high/node 0/affinity 0x0040/b NUL start/high/node 1/affinity 0x0040/b NUL start/high/node 0/affinity 0x0080/b NUL start/high/node 1/affinity 0x0080/b NUL start/high/node 0/affinity 0x0100/b NUL start/high/node 1/affinity 0x0100/b NUL start/high/node 0/affinity 0x0200/b NUL start/high/node 1/affinity 0x0200/b NUL start/high/node 0/affinity 0x0400/b NUL start/high/node 1/affinity 0x0400/b NUL start/high/node 0/affinity 0x0800/b NUL start/high/node 1/affinity 0x0800/b NUL start/high/node 0/affinity 0x1000/b NUL start/high/node 1/affinity 0x1000/b NUL start/high/node 0/affinity 0x2000/b NUL start/high/node 1/affinity 0x2000/b NUL start/high/node 0/affinity 0x4000/b NUL
26. A Principled Technologies test report 26Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 start/high/node 1/affinity 0x4000/b NUL start/high/node 0/affinity 0x8000/b NUL start/high/node 1/affinity 0x8000/b NUL choice/t 20/c yn/d y NUL 2 1 call fg-stream %fff% rem pskill NUL 2 1 pskill NUL 2 1 echo DONE rem End of batch script echo off start/high/node 0/affinity 0x0001/b/w rem End of batch script allow
27. A Principled Technologies test report 27Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 APPENDIX C DETAILED TEST RESULTS Figure 7 is definitely the SPEC CPU2006 ends in detail. Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage Score Out-of-box Optimized Out-of-box Optimized Out-of-box Optimized SPECintrate2006 639 640 618 619 3.4% 3.4% SPECintratebase2006 623 621 593 591 5.1% 5.1% SPECfprate2006 422 422 406 405 3.9% 4.2% SPECfpratebase2006 408 404 399 397 2.3% 1.8% Figure 7: SPEC CPU2006 floating point and integer performance results with the two solutions. Figures 8 and 9 present the detailed outcomes for our LINPACK tests, utilizing the geometric mean at various thread counts. LINPACK results, in GFlops, for out-of-box configurations Number of threads Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage 1 25.5 24.8 2.8% 2 49.9 47.7 4.6% 4 86.0 87.8 - 2.1% 8 147.3 153.6 - 4.1% 16 275.0 249.8 10.1% Figure 8: LINPACK results, in GFlops 10 9 floating-point operations per second, for out-of-box-configurations. LINPACK results, in GFlops, for optimized configurations Number of threads Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage 1 27.2 26.6 2.3% 2 50.9 49.7 2.4% 4 97.4 93.9 3.7% 8 173.3 166.0 4.4% 16 291.9 253.6 15.1% Figure 9: LINPACK results, in GFlops 10 9 floating-point operations per second, for optimized-configurations.
28. A Principled Technologies test report 28Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 Figures 10 and 11 present the detailed outcomes for our STREAM tests. STREAM recent results for out-of-box configurations: Memory bandwidth, in MB/s Number of threads STREAM sub-test Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage 1 Copy 7, 705.1 8, 968.7 - 14.1% Scale 8, 230.0 8, 149.8 1.0% Add 10, 596.2 10, 426.7 1.6% Triad 10, 665.1 10, 481.5 1.8% 2 Copy 17, 476.6 20, 058.6 - 12.9% Scale 18, 442.4 18, 171.8 1.5% Add 23, 530.6 22, 858.2 2.9% Triad 23, 687.6 22, 994.0 3.0% 4 Copy 32, 366.4 24, 843.6 30.3% Scale 34, 544.8 22, 663.2 52.4% Add 44, 124.4 28, 963.6 52.3% Triad 44, 431.2 28, 710.4 54.8% 8 Copy 57, 396.8 44, 924.0 27.8% Scale 61, 013.6 43, 011.2 41.9% Add 67, 405.6 54, 555.2 23.6% Triad 68, 437.6 54, 390.4 25.8% 16 Copy 59, 382.4 69, 147.2 - 14.1% Scale 69, 662.4 61, 038.4 14.1% Add 75, 806.4 74, 046.4 2.4% Triad 66, 822.4 68, 928.0 - 3.1% Figure 10: STREAM memory-performance results with the out-of-box configurations. STREAM latest results for optimized configurations: Memory bandwidth, in MB/s Number of threads STREAM sub-test Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage 1 Copy 7, 662.4 8, 989.6 - 14.8% Scale 8, 848.1 8, 155.1 8.5% Add 10, 790.4 10, 440.6 3.4% Triad 10, 773.3 10, 499.4 2.6% 2 Copy 15, 299.8 20, 138.8 - 24.0% Scale 17, 685.8 18, 247.0 - 3.1%
29. A Principled Technologies test report 29Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 STREAM recent results for optimized configurations: Memory bandwidth, in MB/s Number of threads STREAM sub-test Red Hat Enterprise Linux 6 Microsoft Windows Server 2012 Red Hat advantage Add 21, 571.6 23, 015.2 - 6.3% Triad 21, 532.8 23, 123.four to six.9% 4 Copy 28, 837.6 24, 936.8 15.6% Scale 34, 387.6 22, 372.4 53.7% Add 41, 919.2 29, 046.4 44.3% Triad 41, 649.2 29, 260.0 42.3% 8 Copy 50, 116.8 44, 916.8 11.6% Scale 56, 096.0 43, 095.2 30.2% Add 70, 109.6 55, 236.0 26.9% Triad 75, 199.2 55, 215.2 36.2% 16 Copy 64, 456.0 69, 929.6 - 7.8% Scale 68, 908.8 63, 760.0 8.1% Add 69, 712.0 73, 545.6 - 5.2% Triad 73, 276.8 69, 324.8 5.7% Figure 11: STREAM memory-performance results to the optimized configurations.
30. A Principled Technologies test report 30Comparing CPU and memory performance: Red Hat Enterprise Linux 6 vs. Microsoft Windows Server 2012 ABOUT PRINCIPLED TECHNOLOGIES Principled Technologies, Inc. 1007 Slater Road, Suite 300 Durham, NC, 27703 We provide industry-leading technology assessment and fact-based marketing services. We give every assignment extensive exposure to and expertise in each and every of technology testing and analysis, from researching technology, to developing new methodologies, to testing with existing and new tools. When the assessment is complete, we know the best way to present the final results to a broad range of target audiences. We provide our clients using the materials that they need, from market-focused data to make use of in their own collateral to custom sales aids, including test reports, performance assessments, and white papers. Every document reflects the results of the trusted independent analysis. We provide customized services that give attention to our clients individual requirements. Whether the technology involves hardware, software, Web sites, or services, you can expect the experience, expertise, and tools to assist our clients assess how it can fare against its competition, its performance, its market readiness, and its particular quality and reliability. Our founders, Mark L. Van Name and Bill Catchings, been employed by together in technology assessment more than 20 years. As journalists, they published more than a thousand articles over a wide array of technology subjects. They created and led the Ziff-Davis Benchmark Operation, which developed such industry-standard benchmarks as Ziff Davis Medias Winstone and WebBench. They founded and led eTesting Labs, and as soon as the acquisition of that company by Lionbridge Technologies were the pinnacle and CTO of VeriTest. Principled Technologies can be a registered trademark of Principled Technologies, Inc. All other product names are definitely the trademarks in their respective owners. Disclaimer of Warranties; Limitation of Liability: PRINCIPLED TECHNOLOGIES, INC. HAS MADE REASONABLE EFFORTS TO ENSURE THE ACCURACY AND VALIDITY OF ITS TESTING, HOWEVER, PRINCIPLED TECHNOLOGIES, INC. SPECIFICALLY DISCLAIMS ANY WARRANTY, EXPRESSED OR IMPLIED, RELATING TO THE TEST RESULTS AND ANALYSIS, THEIR ACCURACY, COMPLETENESS OR QUALITY, INCLUDING ANY IMPLIED WARRANTY OF FITNESS FOR ANY PARTICULAR PURPOSE. ALL PERSONS OR ENTITIES RELYING ON THE RESULTS OF ANY TESTING DO SO AT THEIR OWN RISK, AND AGREE THAT PRINCIPLED TECHNOLOGIES, INC., ITS EMPLOYEES AND ITS SUBCONTRACTORS SHALL HAVE NO LIABILITY WHATSOEVER FROM ANY CLAIM OF LOSS OR DAMAGE ON ACCOUNT OF ANY ALLEGED ERROR OR DEFECT IN ANY TESTING PROCEDURE OR RESULT. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC. BE LIABLE FOR INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH ITS TESTING, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC.S LIABILITY, INCLUDING FOR DIRECT DAMAGES, EXCEED THE AMOUNTS PAID IN CONNECTION WITH PRINCIPLED TECHNOLOGIES, INC.S TESTING. CUSTOMERS SOLE AND EXCLUSIVE REMEDIES ARE AS SET FORTH HEREIN.
packages ideal for their machine. The driver packages are named as soon as the kernel they support. Run the
command on the machine to discover your kernel version. The corresponding package provides the name
Phone: 1.800.633.0738