Port Channel Load-Balancing Algorithm (North Bound)

I'm trying to figure out what the load balancing algorithm for the 6100 and 6200 FIs for the Northbound connections. I can't find any documentation on how to change it.
The Nexus 7000s use an 8-bit hash, making it very easy to do something other than 2, 4, or 8 link port channel and get even (at least algorithmically) distribution.
Catalyst switches (not sure about Sup2T though) would use a 1, 2, or 3-bit index, which would skew traffic algorthmically if you used a non-power of 2.
Looking at the 5K documentation, it seems to use the Catalyst style (though haven't been able to confirm). My guess is that whatever is used for the 5Ks is used for the 6100/6200.
Design wise, this would mean you would want to use powers of 2 for your NB uplinks.

Hello Tony,
On UCS FI, it uses " sr-cdest-ip " as the load balancing algorithm and uses 8 parameters for hashing
6248-01-B(nxos)# sh port-channel load-balance
Port Channel Load-Balancing Configuration:
System: source-dest-ip
Port Channel Load-Balancing Addresses Used Per-Protocol:
Non-IP: source-dest-mac
IP: source-dest-ip source-dest-mac
6248-01-B(nxos)# show platform fwm info pc port-channel 1
dump pc info: if_index 369098752 dump_all 0 verbose 1
Po1: state 0x0  #pifs 1  fwimpd ctx 0x9666c1c
Po1: hash params - l2_da 1 l2_sa 1 l3_da 1 l3_sa 1
Po1: hash params - l4_da 1 l4_sa 1 xor_sa_da 1 hash_elect 1
I could not find an option to change these values.
Padma

Similar Messages

  • Nexus 6K: Port-Channel Load-Balance

    Hi all,
    I configured "port-channel load-balance ethernet source-dest-mac" on Nexus 6001. But when I use "show run all | in load-balance", it displays module 1 and module 2 are still using source-dest-ip for port-channel load-balance. And for command "show port-channel load-balance" and "show port-channel load-balance forwarding-path interface", it still shows switch using MAC for hash algorithm. The NXOS is 6.0(2)N1(2a).
    Does anybody know:
    -  What is the function of "port-channel load-balance ethernet source-dest-ip module" and in which situation this command will be effective?
    -  It shows "port-channel load-balance ethernet source-dest-ip module" command for both module 1 and 2. Module 1 is N6K Supervisor and module 2 is 4xQSFP Ethernet Module. Is it possible to set different load-balance algorithm  to these 2 modules?   
    # show run all | in load-balance
    port-channel load-balance ethernet source-dest-mac
    port-channel load-balance ethernet source-dest-ip module 1
    port-channel load-balance ethernet source-dest-ip module 2
    # show port-channel load-balance
    Port Channel Load-Balancing Configuration:
    System: source-dest-mac
    Port Channel Load-Balancing Addresses Used Per-Protocol:
    Non-IP: source-dest-mac
    IP: source-dest-mac
    # show port-channel load-balance forwarding-path interface port-channel 30 vlan 150 src-ip 172.25.228.6 dst-ip 172.25.226.97
    Missing params will be substituted by 0's.
    Load-balance Algorithm on switch: source-dest-mac
    crc_hash: 977 Polynomial: CRC10b        Outgoing port id  Ethernet1/2
    Param(s) used to calculate load-balance:
            seed: 0x701
            dst-mac:  0000.0000.0000
            src-mac:  0000.0000.0000
    # show module
    Mod Ports Module-Type                         Model                  Status
    1   48    Norcal 64 Supervisor                N6K-C6001-64P-SUP      active *
    2   10    Nexus 4xQSFP Ethernet Module        N6K-C6001-M4Q          ok
    Mod  Sw              Hw      World-Wide-Name(s) (WWN)
    1    6.0(2)N2(3)     1.0     --
    2    6.0(2)N2(3)     1.0     --

    Hi all,
    I configured "port-channel load-balance ethernet source-dest-mac" on Nexus 6001. But when I use "show run all | in load-balance", it displays module 1 and module 2 are still using source-dest-ip for port-channel load-balance. And for command "show port-channel load-balance" and "show port-channel load-balance forwarding-path interface", it still shows switch using MAC for hash algorithm. The NXOS is 6.0(2)N1(2a).
    Does anybody know:
    -  What is the function of "port-channel load-balance ethernet source-dest-ip module" and in which situation this command will be effective?
    -  It shows "port-channel load-balance ethernet source-dest-ip module" command for both module 1 and 2. Module 1 is N6K Supervisor and module 2 is 4xQSFP Ethernet Module. Is it possible to set different load-balance algorithm  to these 2 modules?   
    # show run all | in load-balance
    port-channel load-balance ethernet source-dest-mac
    port-channel load-balance ethernet source-dest-ip module 1
    port-channel load-balance ethernet source-dest-ip module 2
    # show port-channel load-balance
    Port Channel Load-Balancing Configuration:
    System: source-dest-mac
    Port Channel Load-Balancing Addresses Used Per-Protocol:
    Non-IP: source-dest-mac
    IP: source-dest-mac
    # show port-channel load-balance forwarding-path interface port-channel 30 vlan 150 src-ip 172.25.228.6 dst-ip 172.25.226.97
    Missing params will be substituted by 0's.
    Load-balance Algorithm on switch: source-dest-mac
    crc_hash: 977 Polynomial: CRC10b        Outgoing port id  Ethernet1/2
    Param(s) used to calculate load-balance:
            seed: 0x701
            dst-mac:  0000.0000.0000
            src-mac:  0000.0000.0000
    # show module
    Mod Ports Module-Type                         Model                  Status
    1   48    Norcal 64 Supervisor                N6K-C6001-64P-SUP      active *
    2   10    Nexus 4xQSFP Ethernet Module        N6K-C6001-M4Q          ok
    Mod  Sw              Hw      World-Wide-Name(s) (WWN)
    1    6.0(2)N2(3)     1.0     --
    2    6.0(2)N2(3)     1.0     --

  • Nexus port channel load balance

    Hi
    I just want to clarify one setting for the port channel load balance on Nexus 6k switch. If I use the load balance option source-dest-ip-only, will following four converstions be load balanced?
    10.10.10.1 -> 192.168.1.1
    10.10.10.2 -> 192.168.1.1
    10.10.10.1 -> 192.168.1.1
    10.10.10.1 -> 192.168.1.2
    Thanks. Leo

    Hi Leo,
    I think there may be typo in your question as I only see three conversations and not four. That aside I've not seen the Nexus port-channel load balancing sufficiently well documented to be able to give you the exact answer.
    In their configuration guides Cisco only include the following statement:
    Cisco NX-OS load balances traffic across all operational interfaces in a port channel by reducing part of the binary pattern formed from the addresses in the frame to a numerical value that selects one of the links in the channel.
    There is other documentation that states the load balancing algorithm uses a CRC-8 based polynomial, but as we don't know exactly which parts of the frame are used in the calculation, I don't see it's possible to calculate the answer and so derive the link that will be used for a given conversation.
    While I've not seen full documentation regarding the science used in the calculation, what Cisco have done is provide a command on the switch CLI that will allow you to determine which link of a port-channel will be used.
    If you run the command show port-channel load-balance forwarding-path interface port-channel vlan src-ip dst-ip then one of the parts of the output is the member link of the port-channel that will be used for that flow.
    You can find full details of the options for the show port-channel load-balance command in the command reference.
    One other point to remember is that the load balancing across a port-channel is unidirectional, and the hashing might be completely different for the return flow of a conversation. For example it is entirely possible that traffic from A to B could use one link of a port-channel, while the return traffic from B to A for the same conversation could use a different link.
    In general I would use the source-dest-port option for load balancing on the Nexus switches as this will obviously include the Layer-4 port numbers in the calculation, and so give you a better distribution of flows across all member links.
    Regards

  • Nexus - port-channel load balancing

    Port-channel   load balancing is a global command or interface command in Nexus switch?
    Thanks,
    Manu

    Hi,
    It's a global command; port-channel load-balance ethernet.
    You can find details in the Configuring Load Balancing Using Port Channels section of the Nexus 5500 Series NX-OS Interfaces Configuration Guide.
    Regards

  • Cisco MDS Port channel load balancing

    A customer recently asked an interesting question about exchange based load balancing on an FC port channel. The platform is UCS with an 8 and 16 port channel per fabric interconnect on two separate UCS domains. The application is Oracles data warehousing which has been known to saturate 4 x 8gb fc links. Since the balancing method is exchange based what constitutes the start and end of an exchange? We are trying to avoid a condition where and intense read write conversation locks to a single link in the port channel and not spread across 8/16 links. Where can I find more information about exchange based routing protocol or how should I go about managing extreme io in a converged infrastructure.

    The default loadbalance method on FI and MDS is src-dst-ox-id based. Note that loadbalancing is done by a device on *outgoing traffic*. FI and MDS do not have to negotiate anything here. Technically, one device can do src-dst-id based while the other can use src-dst-ox-id. However, in your case there is no reason for such a change.
    As I wrote before, If all of the links in the port-channel are touching the max capacity, you should recommend your customer to increase links in the bundle (max 16) or upgrade to 16G links. If few of the links are heavily utilized while other links in the same port-channel are under utilized, you may want to check the application or HBA for capability of breaking down the large reads/writes under smaller exchanges. If no traffic is going on few of the links at all, then I would suspect UCS to FI pinning as well.

  • Port channel Load balancing in Storage VDC

    Hi i am not able to find how to check PO load balancing for storage VDC, although i know by default for FCoE traffic on storage vdc it is OXID but whether it is src-dst l4port or src-dst ipl4port.

    Hi,
    From "Nexus 5500 to Nexus 7000 Multi-Hop FCoE Configuration Example" , 
    Note: On Nexus 7000, by default the source-destination-oxid load balancing mechanism is used for FCoE traffic.
    So let's see what is the default load balancing mechanism in Nexus 7000,
    From "Nexus 7000 interface configuration guide" ,
    The default load-balancing mode for Layer 3 interfaces is the source and destination IP address, and the default load-balancing mode for non-IP interfaces is the source and destination MAC address.
    Which means src-dst ip.
    So what I think is you need to have src-dst ip in default VDC for OXID load balancing in Nexus 7000.

  • Load-balancing Algorithm for NX-OS Port Channels

    Hi, all
    I do not understand description of port-channel load-balance ethernet command.
    switch(config)# port-channel load-balance ethernet ?
      destination-ip         Destination IP address
      destination-mac        Destination MAC address
      destination-port       Destination TCP/UDP port
      source-dest-ip         Source & Destination IP address (includes l2)
      source-dest-ip-only    Source & Destination IP addresses only
      source-dest-mac        Source & Destination MAC address
      source-dest-port       Source & Destination TCP/UDP port (includes l2 and l3)
      source-dest-port-only  Source & Destination TCP/UDP port only
      source-ip              Source IP address
      source-mac             Source MAC address
      source-port            Source TCP/UDP port
    Please tell me what the following descriptions mean.
      Source & Destination IP address (includes l2)
      Source & Destination TCP/UDP port (includes l2 and l3)
    What are the meaning of "includes l2" and "includes l2 and l3" ?
    Thank you for your cooperation in advance.

    Hi Satoru,
    On the Nexus 5000/6000 platforms, all FEXs will inherit the global hashing algorithm from the parent device.
    On the Nexus 7000 platform, hashing algorithms can be assigned on a per FEX basis (all load balancing changes must be made from the Admin VDC):
    N7K-A(config)# port-channel load-balance src-dst ip-l4port fex 134
    Any FEX without a hashing algorithm configured with inherit the global hash. Making changes to the modular/global hash will not alter FEX specific hashing algorithms.
    To verify the configuration applied you can use this command:
    N5K_A# show port-channel load-balance
    On the Nexus 7000, the per FEX algorithm can be checked by appending the ‘fex <#>’ to the end of the command in the Admin VDC or the FEX’s respective VDC:
    N7K-A(config)# show port-channel load-balance fex 134
    Regards,
    Richard

  • 3750X Port-Channel Load-Blanace method

    I have a 3750X and I'm wondering what would be the best Port-Channel Load-Balancing method would be for my network.
    switch(config)#port load ?
      dst-ip       Dst IP Addr
      dst-mac      Dst Mac Addr
      src-dst-ip   Src XOR Dst IP Addr
      src-dst-mac  Src XOR Dst Mac Addr
      src-ip       Src IP Addr
      src-mac      Src Mac Addr
    We have a few Layer 3 VLANs: Default, Servers, Clients, and Guests.
    Some of our servers are LACP bundled.
    So it would be mainly clients on the one layer 3 VLAN accessing the server on the other layer 3 VLAN.
    We also have a few smaller switches that are trunked and LACP bundled back to the core 3750X.
    Can anyone suggest which Load-Balancing method would be best for our situation?
    Thanks!

    The default load balancing method on the 3750 series is based on source-MAC address.  This usually works fine in most cases. If you change it to something different than this, it will affect all your port-channels in that switch.
    HTH

  • ESXi 4.1 NIC Teaming's Load-Balancing Algorithm,Nexus 7000 and UCS

    Hi, Cisco Gurus:
    Please help me in answering the following questions (UCSM 1.4(xx), 2 UCS 6140XP, 2 Nexus 7000, M81KR in B200-M2, No Nexus 1000V, using VMware Distributed Switch:
    Q1. For me to configure vPC on a pair of Nexus 7000, do I have to connect Ethernet Uplink from each Cisco Fabric Interconnect to the 2 Nexus 7000 in a bow-tie fashion? If I connect, say 2 10G ports from Fabric Interconnect 1 to 1 Nexus 7000 and similar connection from FInterconnect 2 to the other Nexus 7000, in this case can I still configure vPC or is it a validated design? If it is, what is the pro and con versus having 2 connections from each FInterconnect to 2 separate Nexus 7000?
    Q2. If vPC is to be configured in Nexus 7000, is it COMPULSORY to configure Port Channel for the 2 Fabric Interconnects using UCSM? I believe it is not. But what is the pro and con of HAVING NO Port Channel within UCS versus HAVING Port Channel when vPC is concerned?
    Q3. if vPC is to be configured in Nexus 7000, I understand there is a limitation on confining to ONLY 1 vSphere NIC Teaming's Load-Balancing Algorithm i.e. Route Based on IP Hash. Is it correct?
    Again, what is the pro and con here with regard to application behaviours when Layer 2 or 3 is concerned? Or what is the BEST PRACTICES?
    I would really appreciate if someone can help me clear these lingering doubts of mine.
    God Bless.
    SiM

    Sim,
    Here are my thoughts without a 1000v in place,
    Q1. For me to configure vPC on a pair of Nexus 7000, do I have to connect Ethernet Uplink from each Cisco Fabric Interconnect to the 2 Nexus 7000 in a bow-tie fashion? If I connect, say 2 10G ports from Fabric Interconnect 1 to 1 Nexus 7000 and similar connection from FInterconnect 2 to the other Nexus 7000, in this case can I still configure vPC or is it a validated design? If it is, what is the pro and con versus having 2 connections from each FInterconnect to 2 separate Nexus 7000?   //Yes, for vPC to UCS the best practice is to bowtie uplink to (2) 7K or 5Ks.
    Q2. If vPC is to be configured in Nexus 7000, is it COMPULSORY to configure Port Channel for the 2 Fabric Interconnects using UCSM? I believe it is not. But what is the pro and con of HAVING NO Port Channel within UCS versus HAVING Port Channel when vPC is concerned? //The port channel will be configured on both the UCSM and the 7K. The pro of a port channel would be both bandwidth and redundancy. vPC would be prefered.
    Q3. if vPC is to be configured in Nexus 7000, I understand there is a limitation on confining to ONLY 1 vSphere NIC Teaming's Load-Balancing Algorithm i.e. Route Based on IP Hash. Is it correct? //Without the 1000v, I always tend to leave to dvSwitch load balence behavior at the default of "route by portID". 
    Again, what is the pro and con here with regard to application behaviours when Layer 2 or 3 is concerned? Or what is the BEST PRACTICES? UCS can perform L2 but Northbound should be performing L3.
    Cheers,
    David Jarzynka

  • Hyper-V NIC Team Load Balancing Algorithm: TranportPorts vs Hyper-VPorts

    Hi, 
    I'm going to need to configure a NIC team for the LAN traffic for a Hyper-V 2012 R2 environment. What is the recommended load balancing algorithm? 
    Some background:
    - The NIC team will deal with LAN traffic (NOT iSCSI storage traffic)
    - I'll set up a converged network. So there'll be a virtual switch on top of this team, which will have vNICs configured for each cluster, live migration and management
    - I'll implement QOS at the virtual switch level (using option -DefaultFlowMinimumBandwidthWeight) and at the vNIC level (using option -MinimumBandwidthWeight)
    - The CSV is set up on an Equallogics cluster. I know that this team is for the LAN so it has nothing to do with the SAN, but this reference will become clear in the next paragraph. 
    Here's where it gets a little confusing. I've checked some of the Equallogics documentation to ensure this environment complies with their requirements as far as storage networking is concerned. However, as part of their presentation the Dell publication
    TR1098-4, recommends creating the LAN NIC team with the TrasportPorts Load Balancing Algorithm. However, in some of the Microsoft resources (i.e. http://technet.microsoft.com/en-us/library/dn550728.aspx), the recommended load balancing algorithm is HyperVPorts.
    Just to add to the confusion, in this Microsoft TechEd presentation, http://www.youtube.com/watch?v=ed7HThAvp7o, the recommendation (at around minute 8:06) is to use dynamic ports algorithm mode. So obviously there are many ways to do this, but which one is
    correct? I spoke with Equallogics support and the rep said that their documentation recommends TransportPorts LB algorithm because that's what they've tested and works. I'm wondering what the response from a Hyper-V expert would be to this question. Anyway,
    any input on this last point would be appreciated.

    Gleb,
    >>See Windows Server 2012 R2 NIC Teaming (LBFO) Deployment and Management  for more
    info
    Thanks for this reference. It seems that I have an older version of this document where there's absolutely
    no mention of the dynamic LBA. Hence my confusion when in the Microsoft TechEd presentation the
    recommendation was to use Dynamic. I almost implemented this environment with switch dependent and Address Hash Distribution because, based on the older version of the document, this combination offered: 
    a) Native teaming for maximum performance and switch diversity is not required; or
    b) Teaming under the Hyper-V switch when an individual VM needs to be able to transmit at rates in excess of what one team member can deliver
    The new version of the document recommends Dynamic over the other two LBA. The analogy that the document
    makes of TCP flows with human speech was really helpful for me to understand what this algorithm is doing. For those who will never read the document, I'm referring to this: 
    "The outbound loads in this mode are dynamically balanced based on the concept of
    flowlets.  Just as human speech has natural breaks at the ends of words and sentences, TCP flows (TCP communication streams) also have naturally
    occurring breaks.  The portion of a TCP flow between two such breaks is referred to as a flowlet.  When the dynamic mode algorithm detects that a flowlet boundary has been encountered, i.e., a break of sufficient length has occurred in the TCP flow,
    the algorithm will opportunistically rebalance the flow to another team member if apropriate.  The algorithm may also periodically rebalance flows that do not contain any flowlets if circumstances require it.    As a result the affinity
    between TCP flow and team member can change at any time as the dynamic balancing algorithm works to balance the workload of the team members. "
    Anyway, this post made my week. You sir are deserving of a beer!

  • SG300-28: LAG Load Balance Algorithm

    The SG300-28 LAG Load Balance Algorithm options are:
    MAC Address
    IP/MAC Address
    This switch is connected to two ESXi hosts running version 5.5 each with quad port 1 gbit nic.
    These ESXI hosts supports > 20 algorithms.
    Is there in a upcoming firmware for the SG300-28 supporting more algorithms?

    The SG300-28 LAG Load Balance Algorithm options are:
    MAC Address
    IP/MAC Address
    This switch is connected to two ESXi hosts running version 5.5 each with quad port 1 gbit nic.
    These ESXI hosts supports > 20 algorithms.
    Is there in a upcoming firmware for the SG300-28 supporting more algorithms?

  • ACE Load Balancing algorithm

    Team,
    I was reading Designing Content Switching Solutions last night. I came across a page that suggested Round Robin for HTTP connections, Least Conns for FTP connections, dst Hash for caching connections and so on.
    Could someone please provide information or a link on which load balancing algorithm to use based on the application, is there some form of best practice for this?
    Thank you,
    John...

    John,
    there is no best practices.
    It depends on your applications and needs.
    For example, for caching, some people prefer to optimize the disk space, and other the response time.
    So, if you do destination hash, you guarantee that all traffic for one site is always handled by the same cache.
    Therefore you optimize the disk space since you will not find the same object on all caches.
    BUT if one site attracts a lot of connections, the cache device that handles that site will be overloaded (for example youtube.com)
    Leastconn is a good option in theory.
    The device that has less connections should receive the next one.
    The problem is if you have flapping links or servers crashing or if you do a lot of maintenance and add/remove servers frequently.
    This confuses the algorithm and is the source of a lot of bugs.
    My recommendation is to go with roundrobin unless you have identify that you really need another algorithm.
    And you can always start with roundrobin and see what happens...
    Gilles.

  • Load balancing algorithm for groups in RZ12

    Hello,
    I would like to know the load balancing algorithm for groups defined in RZ12.
    I know that log on groups for external connections are administered via SMLG and table RZLLICLASS.
    I also know that RFC resources can be managed for RFC logon groups via RZ12.
    Kind regards,
    Peter
    <removed_by_moderator>
    Point awarding is at your discretion, but read and follow the "Rules of Engagement"
    Edited by: Juan Reyes on Dec 3, 2010 10:21 AM

    Hello!
    Found this post while searching information about RFC and Logon Groups...
    I have some mess in my head with SMLG functionality and RZ12. As I know SMLG we can use to distribute users to application server instances, it gives us good achievement in performance. With RZ12 we can distribute RFC connection of particular job for parallel execution on predefined application server. With SMQS and SMQR transaction we can set "Name of AS Group" to route RFC-execution on certain server or servers. But I have troubles with understanding. Imagine, we set up group 1 with 2 servers (name it RFC_GR1), and group 2 with another 2 servers (name it RFC_GR2). How could qRFC scheduler decide on which RFC server group (RFC_GR1 or RFC_GR2) distribute RFC-execution? How to interact "Name of AS Group" with RFC groups if we can set only one group?   How could we distribute RFC-execution depending on our logon groups (smlg)? We would like to distribute RFC depending on SAP logon groups. Is it possible? Or do I compare apple and orange?
    Regards,
    Artem Ivashkin

  • How can ftp service on non-standard port be load balanced using Cisco ACE.

    How can ftp service on non-standard port be load balanced using Cisco ACE.For example ftp service required on tcp 2000 port

    Hi Samarjit,
    you can do this by specifying the port number in the class map that you create . Please find the below mentioend config guide where you can specify the tcp/udp port , range or ports or even the wild card to match the port.
    http://www.cisco.com/en/US/docs/interfaces_modules/services_modules/ace/v3.00_A1/configuration/administration/guide/mapolcy.html#wp1318826
    Regards
    Abijith

  • Load balancing algorithms

    hello
    can any one give URL about load balancing algorithms in 11500 series?
    sincerally

    Hi,
    Here is a link.
    http://www.cisco.com/univercd/cc/td/doc/product/webscale/css/css_730/contlbgd/contrule.htm#1038118
    Hope this helps,
    Regards,
    Sagar

Maybe you are looking for

  • Firefox sync error server incorrectly configured

    The very first time I loaded firefox on my laptop (which i bought only a couple months ago) the syncing worked just fine. After i dont know how much time i started getting errors that the syncing was failing. I tried to reconfigure the syncing proper

  • How to set event name with a apple script

    I use a script to import my photos in iphoto. It's a action folder script. This script automatically create albums with the folder's names. With the latest version of iphoto and ranking per event, I can not initialize the name events with the names o

  • Package Load Failure ODT 11.1.06.20 production

    I have just downloaded the production release of ODT. I uninstalled ODT 11 beta 2. I installed ODT 11 prod without Oracle Services for Microsoft Transaction Server 11.1.6.0. Now I get this error when I want to create a new Oracle Database Project in

  • Cannot open Raw files from RX100M3

    I cannot open (convert?) Raw files from my RX100 M3.   I'm trying to use Photoshop CS5 on a Mac.   I've installed Camera Raw 6.7.1 and Adobe DNG Converter 8.5.

  • New capture bug or am I missing something?

    Problem: I cannot click into or change any of the fields to the right, nor check or uncheck any of the checkboxes or even click on any buttons in the capture window when my Easy Setup is set to "DV-NTSC DV Converter". Some details: I just received an