Скорость интернет в vmware
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
- VMware Technology Network
- :
- Desktop Hypervisor
- :
- VMware Workstation
- :
- VMware Workstation Pro Discussions
- :
- slow guest network performance
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
geeehbee
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The speed on the guest VM is anywhere from 1/10th to 1/2 of the host — no more! On my host I get around 970Mbps down and up — on the guest I will occasionally get a max of 450 down and 350 up — but usually it is in the 120 down and 85 up range. How can I increase the speed on my guest to at least come close to, if not match, the speed on the host.
I’m running VMware® Workstation 12 Pro (12.5.6 build-5528349) and have VMWare Tools installed on the guest. Also, NAT seems to be faster than Bridged.
Host: Win 8.1 64bit, 16Gb RAM, i7 1.7Ghz
Guest: Win 7 64-bit, 12Gb RAM, 1 CPU with 8 cores
I have to run a VPN client in the guest and then the speed really comes down. I will appreciate any help.
gimmely
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
So, you leave only 4GB for the host, when the guest is running. I have to admit that I don’t know if this could have any impact on the network bandwidth/throughput for the guest. But, if the host needs more than 4GB for its OS and any other app when the guest is running, the overall performance should degrade because the host doesn’t have that much in memory left. So, I’d suggest you take a look at memory usage in both the host and the guest and overall performance, with and without the guest running.
geeehbee
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Well, the speeds I mentioned on the host and the guest — they are on the config I mentioned. And as I mentioned, the host has absolutely no issues. So, I think I’m fine with the 4Gb.
Does anyone know of any tweaks I might apply to get the guest network to go faster?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The network performance is ENTIRELY handled by the host CPU. so you absolutely need to leave enough CPU resources for the host to do its work.
geeehbee
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Ok, so is the suggestion that I reduce the number of cores per processor — if so to what amount? And won’t reducing that reduce the overall performance of the VM even more?
wila
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
My suggestion is to start by halving the amount of cores in the guest to 4 cores instead of 8.
And won’t reducing that reduce the overall performance of the VM even more?
nope, usually it makes the guest faster as you lower the CPU scheduling problems.
Frankly unless you run specific software that really takes advantage of multiple cores, once you have 2 cores assigned to the guest, it rarely speeds up things by adding more virtual CPUs to the guest.
A database server might be able to use more cores and things like video rendering, but most workloads don’t benefit from more cores.
| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
geeehbee
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Ok, I created a new VM using converter and assigned it 4 cores. Had to re-enter the Win Serial Number!!
Although, I noticed a minor «average» speed bump (could be placebo though, idk) but not anything significant and still within the initial speeds I mentioned in the post — so still did not get close to the speed on host.
And obviously, when I connect the vpn dialer to tunnel into my remote machine, the speed becomes one tenth of whatever speed the vm had. Essentially, for a 1Gb connection to my host, using NAT in the VM and then VPNing from the guest to a remote machine, I end up with a 24.5 Mbps down, 56 Mbps up connection on the final endpoint Would love to be able to get some more throughput.
As far as provisioning more cores on the guest, well when I’m not dialing in, I’m running device emulators in the VM and those guys will gobble up whatever processing power I can throw.
Anyway, if you all have any other tips, I can try them too but thank you all for your guidance thus far!
Скорость интернет в vmware
- VMWare ESXi 5.0.0-469512-standart хост
- CentOS 6.2 kernel 2.6.32-220.el6.x86_64 — гостевая система в виртуальной локальной сети (так же есть Windows 7 гость)
- CentOS 6.2 kernel 2.6.32-220.el6.x86_64 — гостевая система-маршрутизатор, обеспечивающая выход в глобальную сеть для других гостевых систем
Таким образом создается DMZ внутри одного ESXi хоста.
В такой конфигурации мы замечаем очень медленную работу с centos_test по сети.
Скорость передачи падает чуть ли не до 10 килобайт в секунду.
Диагностика с помощью tcpdump обнаруживает следующее:
IP a.b.c.d > x.x.x.x: ICMP y.y.y.y unreachable - need to frag (mtu 1500), length 556
где a.b.c.d — IP router, x.x.x.x — IP centos_test, y.y.y.y — IP назначения потока данных из глобальной сети.
Проанализировав трафик, понимаем — centos_test пытается отправить пакет заметно большей длины, чем MTU,
на что и получает ответ о необходимости фрагментации.
По каким-то причинам виртуальная сетевая карточка не заботится о максимальном размере пакета, как указано на интерфейсе.
Как решить проблему? Гугл отправляет искать причины в включенном LRO (large receive offload) в ESXi.
Однако никакие рекомендованные манипуляции с настройками (Configuration->Advanced settings->Net), связанными с LRO (Net.VmxnetSwLROSL,Net.Vmxnet3SwLRO,Net.Vmxnet3HwLRO,Net.Vmxnet2SwLRO,Net.Vmxnet2HwLRO и другие),
не дали результата.
Поиски решения дали понять, что в ESXi есть какие-то ошибки в реализации функциональности LRO и TSO (tcp segmentation offload),
которые обнаруживаются в специфичных реализациях. Найти решение на стороне ESXi хоста не удалось, поэтому проблему нужно исправлять на стороне виртуальной машины.
Для этого в гостевой системе нужно отключить поддержку tso (в даном конкретном случае этого достаточно)
на сетевой карточке:
Но в данной конкретной системе (свежая установка Centos 6.2 без апдейта) вместо выключения tso будет ошибка:
Cannot set device tcp segmentation offload settings: Operation not supported
И только после обновления ядра (в моем случае до 2.6.32-220.13.1.el6.x86_64) команда отрабатывает без ошибок.
После этого работа с сетью нормализуется.
Чтобы после перезагрузки tso выключалось без вашего вмешательства, добавьте команду в /etc/rc.local (для CentOS).
В гостевой системе Windows 7 нужно зайти в настройки сетевой карточки и выключить опцию:
“IPv4 — разгрузка большой отправки” для Е1000 и “IPv4 Giant TSO Offload” для vmxnet3.
Скорость интернет в vmware
You are using an outdated browser. Please upgrade your browser to improve your experience.
You can use advanced virtual network adapter settings to limit the bandwidth, specify the acceptable packet loss percentage, and create network latency for incoming and outgoing data transfers for a virtual machine.
Procedure
- Select the virtual machine and select VM > Settings .
- On the Hardware tab, select the virtual network adapter and click Advanced .
- Select a bandwidth setting.
Option Description Limit incoming or outgoing data transfers to the data transfer rate for a specific network connection type Select the network connection type from the Bandwidth drop-down menu. The value in the Kbps text box changes to the data transfer rate, in kilobits per second, of the network connection type that you select. For example, if you select Leased Line T1 (1.544 Mbps) , the value in the Kbps text box changes to 1544. Limit incoming or outgoing data transfers to a specific data transfer rate Select Custom and type the data transfer rate, in kilobits per second, in the Kbps text box. - Enter the acceptable packet loss percentage for incoming and outgoing data transfers in the Packet Loss (%) text box.
The latency setting allows you to simulate the latency in a network environment that differs from your own. The latency range is 0 to 2,000 ms.
Note: Expect actual network latency to be up to 10 ms above the number you set. For example, if you set latency at 200 ms, expect the actual latency to be between 200 to 210 ms.
Скорость интернет в vmware
The Network Rate chart displays network bandwidth on a host.
The Network Data Transmitted/Received chart for hosts is located in the Home view of the Host Performance tab.
- Counter: received
- Stats Type: Rate
- Unit: Megabits per second (Mbps)
- Rollup Type: Average
- Collection Level: 3 (4)
- Counter: transmitted
- Stats Type: Rate
- Unit: Megabits per second (Mbps)
- Rollup Type: Average
- Collection Level: 3 (4)
Chart Analysis
Network performance depends on the application workload and network configuration. Dropped network packets indicate a bottleneck in the network. To determine whether packets are being dropped, use esxtop or the advanced performance charts to examine the droppedTx and droppedRx network counter values.
If packets are being dropped, adjust the virtual machine shares. If packets are not being dropped, check the size of the network packets and the data receive and transfer rates. In general, the larger the network packets, the faster the network speed. When the packet size is large, fewer packets are transferred, which reduces the amount of CPU required to process the data. When network packets are small, more packets are transferred but the network speed is slower because more CPU is required to process the data.
Note: In some instances, large packets might result in a high network latency. To check the network latency, use the VMware AppSpeed performance monitoring application or a third-party application.
If packets are not being dropped and the data receive rate is slow, the host is probably lacking the CPU resources required to handle the load. Check the number of virtual machines assigned to each physical NIC. If necessary, perform load balancing by moving virtual machines to different vSwitches or by adding more NICs to the host. You can also move virtual machines to another host or increase the host CPU or virtual machine CPU.
If you experience network-related performance problems, also consider taking the following actions.
| # | Resolution |
|---|---|
| 1 | Verify that VMware Tools is installed on each virtual machine. |
| 2 | If possible, use vmxnet3 NIC drivers, which are available with VMware Tools. They are optimized for high performance. |
| 3 | If virtual machines running on the same host communicate with each other, connect them to the same vSwitch to avoid transferring packets over the physical network. |
| 4 | Assign each physical NIC to a port group and a vSwitch. |
| 5 | Use separate physical NICs to handle the different traffic streams, such as network packets generated by virtual machines, iSCSI protocols, vMotion tasks. |
| 6 | Ensure that the physical NIC capacity is large enough to handle the network traffic on that vSwitch. If the capacity is not enough, consider using a high-bandwidth physical NIC (10 Gbps). Alternatively, consider moving some virtual machines to a vSwitch with a lighter load or to a new vSwitch. |
| 7 | If packets are being dropped at the vSwitch port, increase the virtual network driver ring buffers where applicable. |
| 8 | Verify that the reported speed and duplex settings for the physical NIC match the hardware expectations and that the hardware is configured to run at its maximum capability. For example, verify that NICs with 1 Gbps are not reset to 100 Mbps because they are connected to an older switch. |
| 9 | Verify that all NICs are running in full duplex mode. Hardware connectivity problems might result in a NIC resetting itself to a lower speed or half duplex mode. |
| 10 | Use vNICs that are TCP Segmentation Offload (TSO)-capable, and verify that TSO-Jumbo Frames are enabled where possible. |