WAN Acceleration Configuration

Hello all,
We have purchased WAAS WAE 674 WAN Accelerator's, and I have a question on placement in our network.  I've just recently implemented redundant WAN Lines (DS3's), and now would like to move my WAE674's so that they are accelerating traffic across both WAN lines.
Is this configuration possible with the 2 WAE674's (no inline cards), and WCCP forwarding, and my HQ WAAS controller, or do I have to purchase an additional 2 WAE 674's to accelarate both WAN lines.
Attached is a drawing for what our WAN design network looks like, we are using EIGRP as our routing protocol, and using per destination load sharing, and redistributing our static routes out from our HQ 6509E.
Please let me know if there is more information needed.
Thanks,
Jon

Hi Jon,
Here are the details of GRE and L2 WCCP redirection.
GRE allows datagrams to be encapsulated into IP packets at the WCCP-enabled router and then redirected to a WAE (the transparent proxy server). At this intermediate destination, the datagrams are decapsulated and then handled by the WAAS software. If the request cannot be handled locally, the origin server may be contacted by the associated WAE to complete the request. In doing so, the trip to the origin server appears to the inner datagrams as one hop. The redirected traffic using GRE usually is referred to as GRE tunnel traffic. With GRE, all redirection is handled by the router software.
Layer 2 redirection is accomplished when a WCCP-enabled router or switch takes advantage of internal switching hardware that either partially or fully implements the WCCP traffic interception and redirection functions at Layer 2. This type of redirection is currently supported only with the Catalyst 6500 series switches and Cisco 7200 and 7600 series routers. With Layer 2 redirection, the first redirected traffic packet is handled by the router software. The rest of the traffic is handled by the router hardware. The branch WAE instructs the router or switch to apply a bit mask to certain packet fields, which in turn provides a mask result or index mapped to the branch WAE in the service group in the form of a mask index address table. The redirection process is accelerated in the switching hardware, making Layer 2 redirection more efficient than Layer 3 GRE.
More details here:
http://www.cisco.com/en/US/docs/app_ntwk_services/waas/waas/v441/configuration/guide/traffic.html
Cisco WAAS Software Release 4.0.13 introduces flexibility when using WCCPv2 as the redirection method. It allows configuration of egress method that increases Cisco WAAS deployment alternatives in cases using WCCP iterception. From Cisco WAAS 4.0.13 onward, the WCCP negotiated return is also supported as the egress method. This method allows the Cisco WAE to be deployed on the same subnet as users or servers and provides better support for preservation of the routing path chosen by the network, because the optimized traffic is returned to the redirecting router. The negotiated return egress method also helps ensure compatibility with asymmetric routing, equal-cost multipath (ECMP) load-balancing, and Hot Standby Router Protocol (HSRP) environments. The return traffic egress method is negotiated based on the WCCPv2 configuration on the router and the egress method configuration on the Cisco WAE.
You will find more information here:
http://www-europe.cisco.com/en/US/prod/collateral/contnetw/ps5680/ps6870/prod_white_paper0900aecd806d976a_ps6474_Products_White_Paper.html
Regards.

Similar Messages

  • WAN Accelerator

    Hi,
    Anyone know if streams will benefit from a wan accelerator?
    The accelerator software will optimize "known protocols and ports" will this include Oracle streams ?, with or without configuration?
    Any info and/or first hand experience is appreciated
    regards Curt

    Depends on where the time is being spent.
    If 90% of the time is spent on the network transmitting data then: Yes. If 90% is spent reading and writing log files and dealing with application logic: No.
    Do you know that the network is the weakest link or are you just shooting at random targets and hoping some arrow sticks?

  • WAN Acceleration

    Hello all. I originally posted this message in the novell.support.native-
    file-access forum and it was suggested that I try this forum instead. So
    here goes.
    We are in the process of evaluating a WAN accelerator solution
    from a company named Riverbed. They have an array of WAN accelerator
    appliances called Steelhead's. These appliances sit at each WAN location
    and actually CACHE the WAN data that is accessed across the wire, thus
    giving your WAN a LAN like performance. We have 3 sites connected via T1
    lines and are a Civil Engineering firm that at times needs to have multiple
    people in multiple offices work on the same projects. We can do this now
    with our setup but the file access performance at the remote sites back to
    our main office is fairly sluggish. The Steelhead appliances are supposed
    to improve this performance dramatically. Spoke to one of their
    representitives and they mentioned that another client of theirs also used
    NetWare and they simply had to enable CIFS on their servers since NCP
    traffic is not accelerated by their devices. I know there is more to the
    configuration than that. Has anyone out there had experience configuring
    NetWare boxes with these Riverbed Steelhead appliances? Any information
    would be appreciated. We are going to be getting some of their reps in
    here soon, hopefully they can shed some light on this as far as what kind
    of configuration they require from their end. We currently have a total
    of 7 NetWare 6.5 SP2 servers spread throughout our 3 locations. We have
    not had a need for any other native file access (CIFS, AFP, NFS) since we
    are a complete Novell shop. As you may have guessed, NFAP is fairly new
    to me, always been a die hard Novell guy. Again, thanks to all for any
    information you could provide.
    John

    Hi
    found some interesting links about your subject. I remember that in my
    company they werde quite impressed about peribit. So go ahead and compare...
    However you have also to consider that theses boxes are not always black
    boxes they mostly need some maintenance....
    www.allot.com
    www.cisco.de
    www.expand.com
    www.nortelnetworks.com
    www.packeteer.com
    www.peribit.com
    www.telonic.de
    Regards
    Marc
    [email protected] wrote:
    > Unfortunately Branch Office doesn't offer bi-directional synchronization
    > which is what most people would look for when deploying somthing like this.
    > On initial setup id does do bi-directional sync, but that's about it.
    >
    > The reason we are looking for a solution like this is because it is one of
    > the very few solutions that offer this type of functionality. Unfortunately
    > these vendors don't offer support for Novell.
    >
    >
    >
    >
    >>We do have native file access (cifs) and consider going back. NCP has
    >>more performance and is more secure. We had abends because of client
    >>security upgrades and were even forced to upgrade from 6.0 to 6.5
    >>rapidly. I always see a higher risk when implementing a service pack
    >>etc. By the way: Why are you going to buy a solution at a vendor with
    >>*NO* (not even little) Novell experience. Consider using Novell Branch
    >>office if you are a real novell shop.
    >>
    >>Bill Sappington wrote:
    >>
    >>>Hey there,
    >>>
    >>>Enabling Windows native file access is just insane, avoid at all costs.
    >>> One of the only major Novell threats came from CIFS, because they of
    >>>course implemented MicroSofts protocol exactly, and therefor implemented
    >>>their bugs and wholes as well. Now one might ask WHY they did this,
    >>>well its because they had to be compatable with a completely exploitable
    >>>file access protocol. They fixed the security whole, and I dont * think
    >>>* it affected the native access protocol for windows, but in my opinion
    >>>windows file access protocols are just not trustable, and well for that
    >>>matter, neither is windows.
    >>>
    >>>Consider BroderManager. It has web caching and all those other goodies,
    >>>including compliant fire walls, client VPN, server to server VPN and all
    >>>that fun stuff and its priced very reasonably. It also lets you control
    >>>who gets to surf, when they get to surf, where they get to surf, etc. etc.
    >>>
    >>>- Bill
    >>>
    >>>[email protected] wrote:
    >>>
    >>>
    >>>>Hello all. I originally posted this message in the
    >>>>novell.support.native-
    >>>>file-access forum and it was suggested that I try this forum instead.
    >>>>So here goes.
    >>>>
    >>>>We are in the process of evaluating a WAN accelerator solution from a
    >>>>company named Riverbed. They have an array of WAN accelerator
    >>>>appliances called Steelhead's. These appliances sit at each WAN
    >>>>location and actually CACHE the WAN data that is accessed across the
    >>>>wire, thus giving your WAN a LAN like performance. We have 3 sites
    >>>>connected via T1 lines and are a Civil Engineering firm that at times
    >>>>needs to have multiple
    >>>>people in multiple offices work on the same projects. We can do this
    >>>>now with our setup but the file access performance at the remote sites
    >>>>back to our main office is fairly sluggish. The Steelhead appliances
    >>>>are supposed to improve this performance dramatically. Spoke to one
    >>>>of their representitives and they mentioned that another client of
    >>>>theirs also used NetWare and they simply had to enable CIFS on their
    >>>>servers since NCP traffic is not accelerated by their devices. I know
    >>>>there is more to the configuration than that. Has anyone out there
    >>>>had experience configuring NetWare boxes with these Riverbed Steelhead
    >>>>appliances? Any information would be appreciated. We are going to be
    >>>>getting some of their reps in here soon, hopefully they can shed some
    >>>>light on this as far as what kind of configuration they require from
    >>>>their end. We currently have a total of 7 NetWare 6.5 SP2 servers
    >>>>spread throughout our 3 locations. We have not had a need for any
    >>>>other native file access (CIFS, AFP, NFS) since we are a complete
    >>>>Novell shop. As you may have guessed, NFAP is fairly new to me,
    >>>>always been a die hard Novell guy. Again, thanks to all for any
    >>>>information you could provide.
    >>>>John
    >>>>
    >>>>
    >
    >

  • Cross-WAN Cluster Configuration in JMQ

    Hi MQ experts,
    Here I have some questions regarding the JMQ cluster configuration
    with regard to the cross WAN support. Let say there are two JMQ
    brokers located in two different geo locations (e.g. one in NY and the
    other in SF) configured to participating in the same JMQ cluster, and
    several JMQ clients are configured to talk to their local JMQ broker in
    either site.
    Given the above configuration, here are my questions:
    * Does this configuration work from the functionality point of view?
    * What is the protocol/mechanism used for the inter-broker communication?
    * Does this configuration (cross WAN) work from the performance's point of view? If so, is there s is any specific configuration that needs to be done (for instance, we heard if HTTP is being used for the data synchronization between these two brokers across the WAN then it's possible, etc.)?
    * For the clients on each site, if the home broker fails, then does the failover mechanism work well in this cross-WAN configuration?
    * If this is indeed a supported configuration, is there any other documentation/performance benchmark reporting from which we can get more information?
    Thanks in advance for any information that you can share with us.
    -Alan

    hi alan,
    1) yes it will work functionally
    2) dont know exactly but i think there are several
    3) what will you use about openmq? what are the charackteristics for your usage? queues/topics temp queues .... lots of things to ask.
    4) it will work
    5) i think it is not supported since WAN capabilities are just in development and are planned for the next releases.
    regards chris

  • WAN port configuration SA520w

    We switched to another internet provider and have problem with our SA520w and a slow WAN connection on av fibre connection.
    We have confirmed that we need to set/ force the WAN port interface on the router to use 10base-t (Full duplex) in order to get maxium speed from our Internet provider. Is there any solution force the router to use that settings as I cant find that alternative on the port setting page in router configuration?
    Latest firmware installed
    Thanks

    Hi,
    You can change speed and duplex settings at Networking->Port Management -> Port Management.
    Thanks,
    Biraja

  • Wan Port configuration

    Hi, I need some info please on how to setup the Wan port.  Here is my setup:  I have 2 locations and 2 UC560.  What I would like to do is beeing able to use one IP on the Wan port in the same subnet as the vlan1 ip address so that I can keep my voice in my tunnel between the 2 offices.  The tunnel is already setup but I would like to avoid having to give a different IP and keep everything similar.
    Is that possible to do?  When I'm trying inside the CCA I can't because it's giving me an error that the IP address is already used by another interface...
    But really I'm giving a different IP.                  
    Any idea without having to do a configuration in the router?
    Thanks
    Jason

    Since I already have the tunnel between the 2 locations I would like to use the existing IP for the wan port so I can do the multisite configuration between the 2 UC560.
    Let say that site1 has a network of 10.10.1.0 and the UC560 VLAN1 has an IP of 10.10.1.5 and site2 has a network of 10.10.2.0 and the UC560 VLAN1 has an IP of 10.10.2.5.  Then I would like having the UC560 of site1 having an IP 10.10.1.8 for the wan port and site2 10.10.2.8.  The problem is that I can't do that under the CCA.  I'm not sure if there is another way to do this but by doing it this way would avoid some configuration in the router where this seem to be always a big deal for our internet provider.
    Let me know if it make sense?
    Thanks again.
    Jason

  • Juniper WAN Acceleration and WCCP v2

    Somewhat off-topic, and perhaps not the right forum anyway, but wccp is part of content networking, isn't it?
    Does anyone have experience of integrating Juniper's WX/WXC products with WCCP ver 2. The reason I ask is that Juniper's documentation says that WCCPv2 is supported, but doesn't seem to give any sense that it really is. Version 2 should support a fault-tolerant configuration where multiple routers operate WCCP with a cluster of caches. According to Cisco IOS documentation, this discovery can be by either of two methods:
    1. Static configuration of the cache IP addresses (i.e. unicast), or
    2. Multicast advertisement
    However, the WX documentation doesn't mention anything about supporting the second (multicast) WCCP discovery method, and doesn't make any explicit reference, either, to supporting multiple statements like
    wccp set router-ip-address 10.0.0.100
    wccp set router-ip-address 10.0.0.101
    to configure the WX to peer WCCP with multiple routers.
    Can anyone advise whether (a) there is any undocumented support for WCCPv2 in multicast mode, or (b) whether the explicit definition of multiple unicast WCCP peer routers is supported?
    I know this should really be a question to put to Juniper's tech support, but I have no access to it and have been asked by a customer to review a proposal including these products.
    Any help greatly appreciated.

    I don't think WCCPv2 is supported

  • SCCM 2012 BITS transfers With Riverbed WAN Acceleration

    Hi all
    i'm implementing SCCM 2012 R2 infrastructure for a organization. 
    its having multiple branch location which i'm planning to put Distribution points and deliver contents using SCCM BITS.
    most of these branch locations is having Riverbed WAN Accelerators, 
    I want to know, BITS working with Riverbed accelerators. is there any issues ? , because i have read in a riverbed document they are saying to disable or have a high bits throttling.
    Please let me know your opinions
    Asitha

    I've seen strange things happening when (misconfigured) WAN accelerators were in place. There are many moving parts involved so there is no one-fits-all answer except for: implement it an try it out.
    Torsten Meringer | http://www.mssccmfaq.de

  • Blue, magenta, magenta, red (power, system, GPS, WAN) light configuration

    Returned from 3 months away 2 days ago and after getting internet reconnected today my network extender will not work. Lights go quickly to Blue, magenta, magenta, red (power, system, GPS, WAN). Have tried restarting several times and have powered down all the devices in the chain, then powered on starting with Cable modem, then Apple Extreme, then network switches, then Network Extender. Still blue, magenta, magenta, red. Any ideas?

    It's connected like it was before I left. I also tried it connected directly to my Apple Extreme wifi box with no difference. The gps is connected directly to the extender. What is odd is that it goes directly to the BMMR lights with none of the expected startup blinking light sequencing.
    Sorry for the response delays. The account is linked to my wife's email, so she has to forward your reponses to me.

  • Solution Accelerator for AP question

    Hi,
    I am working on Solution Accelerator for AP invoice automation. I click on EBSR12 zoom -> process invoices -> login to IPM page -> select an invoice in manual entry queue -> click on Request for invoice information and assign a user from whom I need the information. My question is what happens next. Does the user gets an email asking for information ? Please let me know if you have any ideas.
    Thanks
    Sam

    The answer is that it depends on your accelerator configurations, customization, domain setup, accelerator version, etc. multiple things happen as per your workflow process, but bascally you send the item to the RfI composite, which includes a human task. the HT configuration state what occurs in terms of notification.
    -ryan

  • SA520 Optional port WAN mode for isolated network?

    Long story short, client bought 2 SA520's, one for each site due to it's ability to have a second wan/lan port (Optional Port).  They have a 2nd WAN connection installed that does not have internet.  Esentially it is a hospital link that was install to gain access to some Citrix systems and custom web portals for hispital use. 
    The IDEA was to enable optional port for WAN mode.  Then I was going to add a route statement that any traffic destined to the hospital network would route to the optional wan port,
    Problem - When just enabling WAN port in Optional settings, the WAN port won't come online.  I have to chose load balancing or failover.  Neither of which are true;  I don't want balancing or failover since it isn't a real "internet" connection.
    I thought about setting port to LAN mode but it appears that simply makes the router a 5 port switch instead of 4.  I was hoping I could just simply create a second LAN segment and route to it as well for certain traffic.  Once I enable LAN mode however all configuration disappears and the Networking>LAN menu shows no specific Optional Port configuration options.
    ???  Can this not be configured as such?  Saying it offers a optional port for LAN/WAN/DMZ configurations sort of means doing what I need it to do.  I'm missing something, somewhere.
    Help!
    Eric

    Hi Eric,
    One thing you can do is to enable the Optional WAN port.  Set the 'WAN Mode' to load balancing.  Then use the 'Protocol' Bindings' to handle your routes.  You would configure your Hospital network IP network range as the destination network and select your Optional WAN.  Another rule would be a destination of ANY for the Destination that would go out the dedicated WAN.  You may have to add a rule for DNS resolution to go through your hospital network so the local hospital DNS entries resolve.
    Hope this helps.
    Chris

  • WRV200 PPTP WAN - Default GW is not pointing to tunnel

    Hi all,
    I am currently testing this WRV200.
    We have a PPTP server serving IP range from 192.206.59.1/24
    There is no problem with the authentication and tunnel establishment.
    Just that after the tunnel is established, the default gw in WRV200 is not pointing to the tunnel, but it's still pointing to the gateway on the WAN.
    Since we have chosen the WAN type to be PPTP, the default gateway should be defined on the PPTP tunnel, isn't it?
    Attached is some screenshot.
    Thanks!

    alegalle wrote:I am a liitle confused because of what you are asking and also the screenshots posted.You stated that the WAN is configured for PPTP and it's gateway should be that tunnel. But the screenshot you have of the WRV does not reflec the IP scope of the PPTP server. The highlighted value would be correct as any unknown destination would be sent out the WAN interface. The router also already knows about 192.70.172.0 netwrok and it is attached to its WAN interface.So maybe I did not fully understand your question. From what you have posted it appears that you have a WRTR connecting to the cloud via PPTP and on its LAN we have the WRV connected via its WAN port. In other words like this:CLOUD >> PPTP conn. >> WRTR (WAN: 192.206.59.1 / LAN: 192.70.172.2) >> WRV (WAN *this is the part thats confusing / LAN: 192.168.0.1)I guess I just dont see why the WRV would send all traffic to 192.206.59.0 when it is not a directly attached network. What problems are you encountering with your configuration?
    In PPTP connection, there is no need for the gateway to be in the same subnet.
    Because in the WAN setup screen, we have selected the connection type to PPTP, therefore, the default GW must be pointing to the PPTP gateway.
    The same goes for PPPoE isn't it?
    In the WAN setup screen, if we have selected the connection type PPPoE, the default GW must be pointing to the PPPoE gateway.
    Just that in PPPoE mode, it is usually deployed over copper wires only.
    For PPTP connection, we can deploy it over normal IPv4 network.
    May I know what is the WRTR?

  • Speed limitation on Time Capsule Ethernet WAN port

    I have a Time Capsule 2TB model A1409 that is connected to a Motorola Cable Modem SBV5121 and only connects to it ("green light") when the Ethernet WAN port is set to 10 Mbps/Half Duplex (fastest speed - average 5 Mbps for download) or 10 Mbps/Full Duplex (slowest speed - average 3 Mbps for download). No other mode can get a connection "green light". Time capsule software is 7.6.1.
    Today I connected a Cisco Linksys E2500 router to the same cable modem and download speed increased to 10 Mbps which is what I am paying for my ISP. I haven't even touched the ethernet WAN port configuration. Straight forward.
    I would be grateful if someone could point out a configuration change that would allow my TC to connect as fast as this Cisco router.

    fcamineiro wrote:
    It works fine in Bridge Mode with a Thomson Integrated Modem/Router. But I don't want to use this configuration since I won't be able to use the TC's Guest Network.
    Just tried connecting with Cisco's ethernet cable but remained the same.
    Is it possible to bridge the Thomson??
    Stick a small cheapo switch between the TC and Morotola and see if that fixes it.. as I said, not all chips quite talk to each other.. even in this day and age.
    And Motorola cable modems seem to have more problems than most .. this is not new to me.

  • I wan to start with sql

    i decided to buy ibm server with 48 gb ram and two xeon cpu E5 - 2620 2.0 ghz.
    i wan to make it as RDP Server for 55 user , SQL server with database  (database size is  8 gb) and IIS server(in iis i wan to configure one site which is use data base of sql )
    it is possibale to all these things will manage on my server (ibm server with 48 gb ram and two xeon cpu E5 - 2620 2.0 ghz. ) ?
    if yes then plz tell me how ?
    if no then plz give me proper solutions (What is requirement ?).

     it as RDP Server for 55 user
    RDP for 55 users is not possible only 3-4 users at a timre can connect to windows machine that too also depends on license you take. If 55 users connect to windows machine Via RDP you might get OOM at windows OS level with SQL Server running on it
    Please mark this reply as answer if it solved your issue or vote as helpful if it helped so that other forum members can benefit from it
    My Technet Wiki Article
    MVP

  • I/O Accelerator

    Hi all,
    I see the following in the Cisco MDS 9000 Family I/O Accelerator Configuration Guide
    Limitations and Restrictions:
    IOA clustering framework uses IP connectivity for its internal operation. In Cisco NX-OS Release 4.2(1), if an IOA cluster becomes nonoperational due to IP connectivity, IOA flows are brought down to offline state. In this state, the hosts may not be able to see the targets. To accelerate the IOA flows, the IOA cluster must be operational and there must be at least one IOA switch in each site that is online within this IOA cluster.
    http://www.cisco.com/en/US/docs/switches/datacenter/mds9000/sw/nx-os/configuration/guides/ioa/ioa_config.pdf
    Does this in effect mean that SAN IOA functionality is totally dependent on the reliability of the IP network between the cluster nodes?
    thanks ...

    I don't know if you already got the answer, but I think you can use IPoFC functionality to maintain unified FC and IP traffic for cluster management. However, I've never used interface vsan strategy, so any input will be still welcome :-)

Maybe you are looking for