×

Loading...
Ad by
  • 最优利率和cashback可以申请特批,好信用好收入offer更好。请点链接扫码加微信咨询,Scotiabank -- Nick Zhang 6478812600。
Ad by
  • 最优利率和cashback可以申请特批,好信用好收入offer更好。请点链接扫码加微信咨询,Scotiabank -- Nick Zhang 6478812600。

@

DCI - topic 1 (I)

本文发表在 rolia.net 枫下论坛- virtualization driving 10G at the edge => high density 10G at the edge, 40/100G at the core
- clustered application - vMotion => FrabicPath/TRILL, predictable low latency
- Multi tenants => secure segmentation with VXLAN or LISP/MPLS
- business continuity => L2/L3 DCI connectivity and Storage Extension
- work load mobility => Large L2 Domains such as VXLAN. TRILL/FabricPath

Storage extension

Synchronous data replication impose strict distance between DCs

The application receives acknowledgement for I/O complete when both primary and remote disks are updated. known as zero data loss replication. Speed of light is 300,000 km/s, speed is reduced to 200,000 km/s in fibre, so average 1 ms for light to cross 200 KM of fibre. SCSI protocol (FC) takes four round trips for a write operation => default 50KM, max of 100 KM between two DC's assuming 1 ms latency is required from application with Cisco write acceleration feature on MDS fabric.

Asynchronous data replication
The application receives acknowledgement for I/O complete as soon as primary disk operation is complete. In this mode, there is no distance restriction.更多精彩文章及讨论,请光临枫下论坛 rolia.net
Report

Replies, comments and Discussions:

  • 工作学习 / 科技领域杂谈 / Interoperability between Microsoft and Cisco in collaboration (1)
    本文发表在 rolia.net 枫下论坛Uses cases
    1) Client has an existing MS Instant Messaging and Presence (Lync 2010/2013) deployment, and an existing Cisco voice/video call control environment, how to integrate between the two?
    2) Client has a MS full UC solution (voice, video, IM&P) deployment, in conjunction with an existing Cisco UC environment, how to migrate from one to the other (MS => Cisco of course) ?
    3) This is rather rare, client has full MS UC and full Cisco UC and would like to keep both, how to co-exist?

    New in MS Lync 2013 for UC:
    1) H.264 SVC video codec support (RTVideo in Lync 2010)
    2) Office Web Application Server for PPT sharing
    3) Reduced server roles, mostly consolidated to Front End Server
    4) Enhanced media bypass support (starting from Lync 2010, but would require MTP resources for direct SIP trunk integration as media bypass require a single IP address)


    To provide video end point interoperability:
    1) Cisco VCS server (B2BUA) as video signaling gateway (H264  RTVideo) between MS Front End Pool and Cisco core VCS control, also can work as a proxy to broke enhanced presence status
    2) Cisco AMG to provide HD video transcoding (H263, H264, RTVideo up to 720p@30) between the two islands. Implication: all video media streams will go through this AMG
    3) To enable video end point interoperability between Lync client to Cisco video end points registered on CUCM, 9.x is highly recommended to provide Alpha URI dialing end to end without transformation

    To provide a single call control for voice and video end points:
    1) CuciLync – a Cisco soft client appears to be a plug-in to MS Lync client – have to disable Lync enterprise voice
    2) Support HD video for point to point and multi-point calls

    To provide IM&P federation:
    1) XMPP for IBM Sametime, WebEx Messenger, GTalk,
    2) SIP for Lync
    .......更多精彩文章及讨论,请光临枫下论坛 rolia.net
    • DCI - topic 1 (I)
      本文发表在 rolia.net 枫下论坛- virtualization driving 10G at the edge => high density 10G at the edge, 40/100G at the core
      - clustered application - vMotion => FrabicPath/TRILL, predictable low latency
      - Multi tenants => secure segmentation with VXLAN or LISP/MPLS
      - business continuity => L2/L3 DCI connectivity and Storage Extension
      - work load mobility => Large L2 Domains such as VXLAN. TRILL/FabricPath

      Storage extension

      Synchronous data replication impose strict distance between DCs

      The application receives acknowledgement for I/O complete when both primary and remote disks are updated. known as zero data loss replication. Speed of light is 300,000 km/s, speed is reduced to 200,000 km/s in fibre, so average 1 ms for light to cross 200 KM of fibre. SCSI protocol (FC) takes four round trips for a write operation => default 50KM, max of 100 KM between two DC's assuming 1 ms latency is required from application with Cisco write acceleration feature on MDS fabric.

      Asynchronous data replication
      The application receives acknowledgement for I/O complete as soon as primary disk operation is complete. In this mode, there is no distance restriction.更多精彩文章及讨论,请光临枫下论坛 rolia.net