similar to: Regarding sched-credit weight and caps

I have xen 3.2 running on an 8 core server. I tried to set xm sched-credit -d mydomain --weight=1024 --cap=800 and got: Error: cap is out of range. I thought 8 cores are 800% ?! So, do I have du use 100 (for 100%) instead of 800 (for 100%)? And what about vcpus in my domu config? Must I set vcpus when cpu_weight is set? If yes (e.g. vcpus = 2) does this restrict cpu_weight only to 2 cpus?? Very

Currently with the credit scheduler, cpu_weight cannot be assigned to guest domains at boot time.(it can be set with xm sched-weight after booting) This patch allows CPU weight, and CAP to be set by the Xen configuration files. Specification: The following can be added to the Xen configuration files. 1.CPU WEIGHT [OPTIONAL] cpu_weight = cpu weight value -sets the cpu time ratio to be

Hello Everyone, I am hoping I can get some help with how vCPUs work with weights and caps 1. If I have a domU with 2 vCPUs and a weight of 256, does the weight double (512) or is this 256 time slice across both vCPUs. If the weight does not double, how is it beneficial to have multiple vCPUs? 2. If the domU has 2vCPUs and a cap of 100% how does this work since 100% is a mex of one physical

Hi Everyone, How can I make use of xm sched-credit in a VPS hosting environment? Can I use "xm sched-credit -d <domain>" in a script at system startup, or does this have to be run after the respective guest has been started? Also, what happens when the guest is shutdown then restarted, or rebooted? Do I need to run that command again? If so, how would I automate this process?

Modified function runq_candidate in the credit 2 scheduler to have it consider hard and soft affinity when choosing the next vCPU from the run queue to run on the given pCPU. Function now chooses the vCPU with the most credit that has hard affinity and maybe soft affinity for the given pCPU. If it does not have soft affinity and there is another vCPU that prefers to run on the given pCPU, then as

Wondering if I can get some opinions on dom0 CPU pinning and xm weighting. We have several xen servers on SLES11SP1 and up to this point we have never done anything with pinning CPU''s to dom0. We''re trying to decide if we should be doing that. Same goes for xm weighting (if I''m saying that correctly). For example, running the following command when a xen server

Hi Everyone, What''s the parameter in the DomU config file to cap the CPU on the domain? I tried using "cap = 20". While the VM booted ok, doing a "xm sched-credit -d </domain/>" showed the cap as being 0 Any ideas? Thanks _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com

On a 8 Core system only dom0 can set cap > 100 and vcpus > 1. On a domU ''xm sched-credit -d <domU> -c 200'' says "Error: cap is out of range". Vcpu cannot be set higher than 1 - regardless of boot/config/command line. Low performance is equivalent to the cap setting. See ''xm list --long'': (domain (domid 0) (uuid

Hi all, The xen documentation at http://wiki.xensource.com/xenwiki/CreditSchedulerstates that the max weight you can set for a domU is 65535. To verify this I decided to try setting 65535 to my dom0 which had no domUs running at the time. I was unable to set a weight of 65535. The server is a duel quad core (8 cores) with 32GB RAM running Centos 5.4/xen-3.0.3-94.el5_4.3 . The maximum weight I was

Implements tool interfaces for scheduling domains. libxenctrl, xm, and xend. signed-off-by: Mike D. Day <ncmike@us.ibm.com> -- libxc/xc_domain.c | 85 +++++++++++++++++++++++++++++++++--- libxc/xenctrl.h | 43 ++++++++++++++++-- python/xen/xend/XendDomain.py | 78 +++++++++++++++++++++++++++++++++ python/xen/xend/server/SrvDomain.py |

Hi all, Since Xen 4.2.0, users can change time slice of the scheduler at runtime via xl command line, a very nice feature. However, it is not *correctly* implemented. Problem description -------------------- say you set the ''cap'' of one VM to 50 (a half core), -when setting the time slice to be *greater* than 30ms, the VM gets much *less* CPU cycles than its allocation -when

Hi all (and Mr. Dunlap in particular), I have a question about the credit (and ultimately credit2) scheduler that I hope you can help me with. I have read the white paper "Scheduler development update" and as much material on the credit scheduler as I can find, but I am still not completely clear on how I should think about the cap. Example scenario: Server hardware: 2

Hi all! I have a doubt about shared resources: If I have a Dom-0 and two virtual machines Dom1 and Dom2, how can I split cpu resources? I mean how can I assign exactly 25% of cpu to Dom1 and 75% of cpu to Dom1? (I assume that Dom-0 uses 0% of cpu). Can I also choose which is the virtual machine I assign the disk I/O speed? Thanks very much! Marco

I''d like to implement something like SFQ, but with a cap of a rate per flow. Essentially, I''d like to be able to limit citrix connections to 30KBps without having an overall traffic flow cap for all citrix connections. Slightly preferable to this would be something that would actually work per flow instead of the stochastic algorithm. Is this possible? Thanks, Bill --

Dear List, I couldn't find any package that performs the "weight of evidence" of predictors (a transformation usually performed in credit scoring applications). Is there any that you know? Thanks, Axel. [[alternative HTML version deleted]]

Hi Is there a doc/wiki page which explains, the kind of domu images/appliances/kernels can be run with what type of dom0 kernels(xen_caps). For example : -- Can a 32 bit, non-PAE domU run on 64 bit, PAE dom 0 in PV mode ? (or HVM mode ?) Are there any checks that can be done before doing the migration ? Is there any programmatic way (from python) to find out if the kernel in a domU disk

This patch -Displays[1] current parameters for running domains ala xm list -Allow users to set[2] one or more parameters[3] without having to provide values for parameters they do not wish to change -Adds additional testing of sched-sedf via new xm-test testcases. With this patch applied, test 02_sedf_period_lower_neg.py exposes a bug[4]. I''ll follow up this email with a patch for

Good day. I found that XCP ignores VCPU-params (weight and cap) after VM live migration, as well, as those parameters ignored if changed on running VM. May I as someone with xcp cloud nearly check this bug? Thank you. -- wBR,George. _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users

As discussed in a previous e-mail, this patch series implements yield for the credit scheduler. This allows a VM to actually yield (give up the cpu to another VM) when it wants to. This has been shown to be effective when used in the spinlock code to avoid wasting time spinning when another vcpu is not currently scheduled. _______________________________________________ Xen-devel mailing list

Hi! I`m using xen 3.0.2 on a Debian Sarge 3,1 (XEON machine)! By default my scheduler is "sedf", ok?! I would like to change it to the "credit"! I tried to set the parameter "sched=credit" in my grub config, but when I try to use the command "xm sched-credit -d <domain>" and others, it doesn`t work! But if I try "xm sched-sedf -l" it