Jump to content
Калькуляторы

Управление прерываниями

Reply to this topic

foxroot   

foxroot
Posted · Report post

Добрый день!

Есть сервер HP 180 G6

 

Хочу сделать распределение нагрузки на ядра процессора

подскажите как правильно раскидать

 

сейчас вот такая картина

top - 18:10:00 up 4:16, 1 user, load average: 0.00, 0.06, 0.04

Tasks: 217 total, 1 running, 216 sleeping, 0 stopped, 0 zombie

Cpu0 : 0.3%us, 0.7%sy, 0.0%ni, 87.9%id, 0.0%wa, 0.0%hi, 11.1%si, 0.0%st

Cpu1 : 0.3%us, 0.3%sy, 0.0%ni, 63.6%id, 0.0%wa, 0.0%hi, 35.7%si, 0.0%st

Cpu2 : 0.3%us, 0.3%sy, 0.0%ni, 77.9%id, 0.0%wa, 0.0%hi, 21.5%si, 0.0%st

Cpu3 : 0.7%us, 0.3%sy, 0.0%ni, 87.0%id, 0.7%wa, 0.0%hi, 11.3%si, 0.0%st

Cpu4 : 0.3%us, 0.3%sy, 0.0%ni, 99.3%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st

Cpu5 : 0.3%us, 0.7%sy, 0.0%ni, 85.1%id, 0.0%wa, 0.0%hi, 13.9%si, 0.0%st

Cpu6 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st

Cpu7 : 0.0%us, 1.3%sy, 0.0%ni, 98.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st

Mem: 5981756k total, 1661796k used, 4319960k free, 40240k buffers

Swap: 6142972k total, 0k used, 6142972k free, 741008k cached

 

недогруженный CP4 CP6 CP7

 

в основном трафик идет eth5

cat /proc/interrupts

CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7

 

95: 22677190 15757231 9052251 18089402 27271540 31185129 6854059 4526608 PCI-MSI-edge eth5-TxRx-0

96: 3973082 6962279 12611155 13913986 22203411 11676210 26678511 39955028 PCI-MSI-edge eth5-TxRx-1

97: 32063312 14400603 9270278 10412950 19489226 21945577 18313258 11608129 PCI-MSI-edge eth5-TxRx-2

98: 12654337 27723466 13306100 9918579 12994023 16664310 12068137 35291864 PCI-MSI-edge eth5-TxRx-3

99: 19963845 26226626 30314161 14684495 12820147 15195722 4795258 17395439 PCI-MSI-edge eth5-TxRx-4

100: 10528312 12028815 23002907 42778657 18815283 11708788 203862 19641374 PCI-MSI-edge eth5-TxRx-5

101: 3557727 4962843 5669170 13574142 44765439 48317876 17928494 2342587 PCI-MSI-edge eth5-TxRx-6

102: 515086 140830 206850 2116227 10333245 12019618 48600542 66612864 PCI-MSI-edge eth5-TxRx-7

Share this post

Link to post
Share on other sites

NiTr0   

NiTr0
Posted · Report post

подскажите как правильно раскидать

прибить очереди к ядрам.

 

к слову, ядер-то у вас всего 4, и возможно отключение гипертрединга уменьшит нагрузку...

Share this post

Link to post
Share on other sites

hsvt   

hsvt
Posted (edited) · Report post

Интересует тот же вопрос, вот конфиг:

 

lspci | grep -i ethernet
01:00.0 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
01:00.1 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
03:00.0 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
03:00.1 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)

 

driver: igb
version: 5.3.4.4
firmware-version: 1.2.1

 

model name: Intel(R) Xeon(R) CPU L5639  @ 2.13GHz
grep  ^processor /proc/cpuinfo
processor       : 0
processor       : 1
processor       : 2
processor       : 3
processor       : 4
processor       : 5
processor       : 6
processor       : 7
processor       : 8
processor       : 9
processor       : 10
processor       : 11

 

cat /proc/interrupts | grep eth
 64:          2          0          0          0          0          0          0          0          0          0          0          0   PCI-MSI-edge      eth0
 65:         78          0          0          0          0          0          0          0          0          0        510          0   PCI-MSI-edge      eth0-TxRx-0
 67:          2          0          0          0          0          0          0          0          0          0          0          0   PCI-MSI-edge      eth1
 68:        115          0          0          0          0          0          0          0        379          0          0          0   PCI-MSI-edge      eth1-TxRx-0
 69:          2          0          0          0          0          0          0          0          0          0          0          0   PCI-MSI-edge      eth2
 70:         33          0          0          0          0          0        212          0          0          0          0          0   PCI-MSI-edge      eth2-TxRx-0
 71:          2          0          0          0          0          0          0          0          0          0          0          0   PCI-MSI-edge      eth3
 72:        188          0          0          0       2673          0          0          0          0          0          0          0   PCI-MSI-edge      eth3-TxRx-0

 

dmesg | grep igb | grep -v MAC
[    3.170035] igb 0000:01:00.0: setting latency timer to 64
[    3.175223] igb 0000:01:00.0: irq 64 for MSI/MSI-X
[    3.175231] igb 0000:01:00.0: irq 65 for MSI/MSI-X
[    3.366060] igb 0000:01:00.0: added PHC on eth0
[    3.366153] igb 0000:01:00.0: Intel(R) Gigabit Ethernet Network Connection
[    3.366252] igb 0000:01:00.0: eth0: (PCIe:2.5GT/s:Width x4)
[    3.366560] igb 0000:01:00.0: eth0: PBA No: 82576B-001
[    3.366652] igb 0000:01:00.0: LRO is disabled
[    3.366742] igb 0000:01:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)
[    3.366909] igb 0000:01:00.1: setting latency timer to 64
[    3.367190] igb 0000:01:00.1: irq 67 for MSI/MSI-X
[    3.367197] igb 0000:01:00.1: irq 68 for MSI/MSI-X
[    3.567791] igb 0000:01:00.1: added PHC on eth1
[    3.567883] igb 0000:01:00.1: Intel(R) Gigabit Ethernet Network Connection
[    3.567982] igb 0000:01:00.1: eth1: (PCIe:2.5GT/s:Width x4)
[    3.568291] igb 0000:01:00.1: eth1: PBA No: 82576B-001
[    3.568383] igb 0000:01:00.1: LRO is disabled
[    3.568472] igb 0000:01:00.1: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)
[    3.568628] igb 0000:03:00.0: setting latency timer to 64
[    3.568961] igb 0000:03:00.0: irq 69 for MSI/MSI-X
[    3.568967] igb 0000:03:00.0: irq 70 for MSI/MSI-X
[    3.759826] igb 0000:03:00.0: added PHC on eth2
[    3.759917] igb 0000:03:00.0: Intel(R) Gigabit Ethernet Network Connection
[    3.760015] igb 0000:03:00.0: eth2: (PCIe:2.5GT/s:Width x1)
[    3.760328] igb 0000:03:00.0: eth2: PBA No: E43709-003
[    3.760419] igb 0000:03:00.0: LRO is disabled
[    3.760509] igb 0000:03:00.0: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)
[    3.760666] igb 0000:03:00.1: setting latency timer to 64
[    3.760999] igb 0000:03:00.1: irq 71 for MSI/MSI-X
[    3.761007] igb 0000:03:00.1: irq 72 for MSI/MSI-X
[    3.951874] igb 0000:03:00.1: added PHC on eth3
[    3.951966] igb 0000:03:00.1: Intel(R) Gigabit Ethernet Network Connection
[    3.952064] igb 0000:03:00.1: eth3: (PCIe:2.5GT/s:Width x1)
[    3.952377] igb 0000:03:00.1: eth3: PBA No: E43709-003
[    3.952468] igb 0000:03:00.1: LRO is disabled
[    3.952558] igb 0000:03:00.1: Using MSI-X interrupts. 1 rx queue(s), 1 tx queue(s)
[   12.389764] igb 0000:01:00.0: changing MTU from 1500 to 9000
[   12.581777] igb 0000:01:00.1: changing MTU from 1500 to 9000
[   12.774347] igb 0000:03:00.0: changing MTU from 1500 to 9000
[   12.966353] igb 0000:03:00.1: changing MTU from 1500 to 9000
[   14.567936] igb 0000:01:00.0: eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   14.667840] igb 0000:01:00.1: eth1: igb: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   14.851953] igb 0000:03:00.0: eth2: igb: eth2 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   15.059960] igb 0000:03:00.1: eth3: igb: eth3 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   23.800159] igb 0000:01:00.0: eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   24.040158] igb 0000:01:00.1: eth1: igb: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   24.184275] igb 0000:03:00.0: eth2: igb: eth2 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   24.316277] igb 0000:03:00.1: eth3: igb: eth3 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

 

82576 должна создавать по идее по 8 очередей же ? А здесь же всего по 1 очереди от каждой с совмещёнными векторами.

Edited by hsvt

Share this post

Link to post
Share on other sites

snvoronkov   

snvoronkov
Posted · Report post

82576 должна создавать по идее по 8 очередей же ? А здесь же всего по 1 очереди от каждой с совмещёнными векторами.

http://forum.nag.ru/forum/index.php?showtopic=81447

Share this post

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
Followers 0
Go to topic listing Программное обеспечение, биллинг и *unix системы