前5个sip - truntchas !

2009年，Nemertes采访了大约200名IT高管，以收集有关他们IT计划的数据。我们的目标之一是将SIP集群的现实与炒作分开。我们的发现证实了我们最初的假设:对SIP集群的兴趣正在增长，早期的采集者正在经历成功，但许多人也面临着挑战。我们刚刚结束了另一轮的采访，虽然我们仍处于分析数据的早期阶段，但我们已经确定了一些实现SIP集群的人所确定的关键“陷阱”。1.互操作性。我们一次又一次地听到人们对互操作性的担忧。尽管有SIPconnect 1.0标准，尽管供应商承诺互操作性和许多认证程序，采用者仍然告诉我们，他们最大的问题只是克服了当前pbx和SIP集群提供商之间的互操作性问题。通常情况下，在开始实现后，他们会发现自己需要软件升级、网关或会话边界控制器。我们听说过对SIP接口的各种支持; some providers deploy an integrated access device that converts their own SIP service to a TDM interface going to the customer PBX. Others support direct SIP connectivity (perhaps with an SBC for security). In some cases the choice of provider architecture breaks features such as the ability for a local IP-PBX to fail over to a remote server. 2. Applications and Legacy services. Common concerns heard include the inability of SIP trunking services to support fax, dial-up modems, and call recording software or other customized contact center applications. Again these issues tend to rear their heads after deployments have started. E.911 service support isn’t widely available as well. 3. Architecture. Those seeing the biggest benefits from SIP trunking don’t simply replace their existing PSTN trunks at all locations with an equal number of SIP trunks; rather they deploy SIP trunking as part of a move to a centralized architecture. This approach allows architects to reduce the number of trunks, take advantage of bursting services, and more easily support distribution of calls into multiple data centers. A centralization strategy often entails deployment of session management or some other means of unifying disparate systems. 4. Management. We still hear concerns around performance management. Vendors and service providers are still coming up to speed in their ability to troubleshoot and proactively manage SIP trunking services, as well as their ability to enforce SLAs covering up-time, call completion and call quality. 5. Design. In a one-for-one replacement of TDM PSTN trunks with SIP trunks, bandwidth costs for SIP trunking actually exceeded TDM costs due to the additional overhead required for SIP compared with 64 Kbps channels for TDM voice. Growing use of high quality compression algorithms such as ITU G.722 and G.729 will allow SIP trunking adopters to reduce bandwidth requirements. Despite all these concerns I expect that our data will show SIP trunking interest is continuing to grow. Those of you responsible for telecom strategy should continue to evaluate SIP trunking services, especially as part of your overall WAN and mobility strategy, but pay special attention to the concerns I raised in this list; doing so will save you a great deal of headaches down the road.