I'm working on setting up a test environment, but I need to scale down the hardware we're using for our SQL box, which is running Windows Server 2008 R2 SP 1 and SQL 2008. I'm noticing that MSConfig.exe has options for limiting the number of CPUs available, but I'm not able to find any documentation on how that works on MSDN.
I'm also not seeing any way to change this via command line using the BCDEdit that Microsoft put in.
Anyone know of documentation on this? I'm trying to decide whether limiting the Processor usage at boot would be the best test, or limiting the processors in SQL itself. I'm leaning towards boot time because I'm trying to accurately mimic a lower-power physical box, and if I limit the power at the database level the extra power may show up in other areas.
In an era of multi-core, hyperthreaded CPUs, "Processors" is now an ambiguous term. Does the processor refer to a threads (from Hyperthreading), cores, or sockets (physical CPUs).
Windows recognises logical processors (LP) as the basic compute platform - giving one LP for each hyperthread within a core, and then multiplied for each core, which is finally multiplied by the number of sockets.
The easiest way in Windows to reduce the LP count is to use the /NUMPROC option. In this example itc creates a maximum number of processors of 8.
Backup bcdedit /export c:\Backup\bcd.bak
List of current entries bcdedit /v
Copy of existing config bcdedit /copy {current} /d "Windows 2008 R2 with NumProc"
The current is a "well known" identifier . This also returns the Id that is created. So that can be used directly
List of entries bcdedit /v
adding parameter to new entry To set maximum number of processors
bcdedit /set {new_ID} NUMPROC 8
Changing default entry bcdedit /default {new_ID}
But the danger of this strategy is for example on a 4 core system with HyperThreading enabled then only one processor (socket) is used. Since the first 8 LPs (0..7) will be hyperthreaded cores on the first processor. So you are not really emulating an 8 way system - but a 1 way - 4 core system with HyperThreading enabled.
If this doesn't fit your model other options are
Disable HyperThreading in the BIOS - this reduces the LP count per core so helping distribute the load over cores and sockets.
Does the System BIOS support reducing the core counter per processor? If so this will help distribute loads over sockets.
Building your system within a virtual environment and limiting physical resources from that perspective.
If you are dealing with more than 64 Logical Processors under Windows then that introduces Processor Groups and that add another layer of options.
Related
I am using PostgreSQL 9.5 with a Java 8 Application on Windows OS (System: i5 2nd Generation). I noticed that when my application is in execution, there are several PostgreSQL processes / sub-processes that are created/removed dynamically.
These PostgreSQL processes use almost all of the CPU (>95%), due to which problem arises with other applications installed at my system.
I recently came to know about CPU affinity. For the time being, I am executing PowerShell script (outside of my Java application) which checks periodically and sets the desired value of cpu affinity for all PostgreSQL processes in execution.
I am looking for a way where I don't need to execute an external script(s) and/or there is one-time configuration required.
Is there a configuration supported by PostgreSQL 9.5 through which we can set max CPU cores to be used by PostgreSQL processes?
I looked for the solution, but could not find any.
There is no way to set this in the PostgreSQL configuration.
But you can start your PostgreSQL server from cmd.exe with:
start /affinity 3 C:\path\to\postgresql.exe C:\path\to\data\directory
That would allow PostgreSQL to run only on the twp “first” cores.
The cores are numbered 1, 2, 4, 8, 16 and so on, and you use the sum of the cores where you want PostgreSQL to run as argument to /affinity. For example, if you only want it to run on the third and fourth core, you would use /affinity 12.
This should work, since the Microsoft documentation says:
Process affinity is inherited by any child process or newly instantiated local process.
i have question about KVM i could not find any satisfying answer in the net about.
Lets say i want to create 3 virtual machines on a host with 2 CPUs. I am assigning 1 cpu to 1 virtual machines. The other 2 virtual machines should be sharing 1 cpu. If it is possible i want to give 1 vm 30 % and the other one 70% of the cpu.
I know this does not make much sense but i am curious and want to test is :-)
I know that hypervisors like onapp can do that. But how do they do it?
KVM represents each virtual CPU as a thread in the host Linux system, actually as a thread in the QEMU process. So scheduling of guest VCPUs is controlled by the Linux scheduler.
On Linux, you can use taskset to force specific threads onto specific CPUs. So that will let you assign one VCPU to one physical CPU and two VCPUs to another. See, for example, https://groups.google.com/forum/#!topic/linuxkernelnewbies/qs5IiIA4xnw.
As far as controlling what percent of the CPU each VM gets, Linux has several scheduling policies available, but I'm not familiar with them. Any information you can find on how to control scheduling of Linux processes will apply to KVM.
The answers to this question may help: https://serverfault.com/questions/313333/kvm-and-virtual-to-physical-cpu-mapping. (Also that forum may be a better place for this question, since this one is intended for programming questions.)
If you search for "KVM virtual CPU scheduling" and "Linux CPU scheduling" (without the quotes), you should find plenty of additional information.
I am using qemu on Fedora and I find that qemu does not support multi-core. When I use the parameter smp and set cores=2, it will tell me that:
mu-system-riscv: Number of SMP CPUs requested (2) exceeds max CPUs supported by machine 'riscv'
In general, QEMU can support multicore guests, yes. However the number of cores supported depends on the particular board (machine) model you're using. The error message is telling you that the 'riscv' machine you've asked for only supports one CPU.
(In TCG emulation at the moment multicore guests won't be any faster than a single core guest because we don't use all the host cores; this should change in QEMU 2.9 for at least some host/guest combinations when multithreaded TCG support lands. KVM supports multicore guests with no problems.)
I was looking for an answear but didn't find one.
I'm trying to create a new VM to develop a web application. What would be the optimal processor settings?
I have i7 (6th gen) with hyperthreading.
Host OS: Windows 10. Guest OS: CentOS.
Off topic: RAM that should I give to VM should be 50% of my memory? Would it be ok? (I have 16GB RAM)
Thanks!
This is referred to as 'right-sizing' a vm, and it is dependent on the application workload that will run inside it. Ideally, you want to provide the VM with the minimum amount of resources the app requires to run correctly. "Correctly" is subjective based upon your expectations.
Inside your VM (CentOS) you can run top to see how much memory and cpu % is being used. You can also install htop which you may find friendlier than top.
RAM
If you see a low % of RAM being used, you can probably reduce what you're giving the VM. If you are seeing any swap memory used (paging to disk), you may want to increase the RAM. Start with 2GB and see how the app behaves.
CPU
You'll may want to start with no more than 2vCPUs, checkout top to see how utilized the application is under load, and then make an assessment for more/less vCPUs.
The way a hosted hypervisor (VMware Workstation) handles guest CPU usage is through a CPU scheduler. When you give a vm x number of vCPUs, the VM will need to wait till that many cores are free on the CPU to do 'work'. The more vCPUs you give it, the more difficult (slower) it will be to schedule. It's more complicated than this, but I'm trying to keep it high level. CPU scheduling deep dive.
I am trying to understand hardware assisted virtualization for a project with ARM CortexA8 and using the ARM Trustzone feature. I am new to this topic therefore I started with Wiki entries to understand more.
Wikipedia explains hardware assisted virtialization and adds a line in the definitionas:
Full virtualization is used to simulate a complete hardware
environment, or virtual machine, in which an unmodified guest
operating system (using the same instruction set as the host machine)
executes in complete isolation.
The text in bold is a bit confusing. How is the same instruction set of the processor used to provide two isolated environment? Can someone explain it? ArmTrustzone manual also talk of a "virtual processor core" to provide security. Please throw some light.
thanks
The phrase "using the same instruction set as the host machine" means that the guest OS is not aware of the virtualization layer and behaves as if it is executed on a real machine (with the same instruction set). This is in contrast to the para-virtualization paradigm in which the guest OS is aware of virtualization and calls some specific VMM functions, i.e. hypercalls.
No, CPU has not additional instructions. Virtual machine instruction set is translated by a hypervisor component called VMM (virtual machine manager) to be executed on the physical CPU.
Physical CPU with assisted Virtualization introduced only a new ring 0 mode called VMX that allow the virtual machine to execute some instructions in ring 0.