We think sipXecs is the leading open source IP PBX. It excels in terms of scalability, robustness and ease of use. The sipXecs IP PBX has been successfully deployed in thousands of places. The largest known installation serves more than 6,000 users connected to one redundant (HA) system. Small installations go all the way down to a few users served by very low cost hardware. sipXecs is a very sophisticated VoIP unified communications system. The fact that it still is so incredibly easy to use is probably its single most important accomplishment.

 There is a detailed list of features on the sipX Wiki (it is too long to re-produce here). Here is why you should consider using sipXecs:

1. You need an IP PBX that is robust, stable and easy to use
2. Need a system that offers plug & play management of all the components including phones
3. It is a mission critical system and therefore you insist in high-availability redundant configuration. A server failure should not interrupt calls
4.  Voice quality is important to you and therefore you want a system that routes media peer-to-peer and not through the PBX
5. SIP standards compliance that allows flexible call routing including SIP trunking
6. Need scalability - sipXecs is known to work for deployments in excess of 5,000 users connected to one redundant server
7. Lots of features - of course, who does not want that
8. Unlimited number of simultaneous calls, and unlimited number of trunks using distributed gateways
9. Need IT integration with no special needs for specific HW. Runs on off-the-shelf Intel servers
10. You run other Microsoft applications like Active Directory and Exchange and want your IP PBX to integrate with them
    

The project roadmap is updated regularly also on the sipX Wiki. We solicit input from the community to align priorities for every month of development. We aim for a release interval of 6 to 8 month between major releases.

Key capabilities of sipXecs include:

• Fully featured Unified Communications system: sipXecs is a complete IP PBX applications that has all the features you would expect from a unified communications solution.
• Ease of installation:  sipXecs installs in hours, guaranteed. It runs on any standard server without the need for any special hardware.
• Easy to use: No need to get expert help. With sipXecs you will be self-sufficient for all adds, moves and changes. Typically a receptionist is capable of managing the system. There are no hidden configuration files or other things that requre a specialist.
• Plug & play management of everything: One - two - three clicks and you just configured a new user with a phone. The phones are auto-discovered and as soon as they are connected to the LAN pick up configuration from sipXecs and come up configured. No messing around with phone or gateway configuration ever again.
• All features included:  For the 3.10 release alone we added more than 170 new features.
  
• Redundancy and scalability: sipXecs is unique in that it offers full load-sharing redundancy for the call control system. A server failure will not cause calls to be interrupted. sipXecs is architected as a distributed system. It scales by simply adding hardware. Need more capacity for your call center ACD application?  Run it on separate hareware or add an additional ACD server.
• Trunk redundancy and failover: sipXecs uses external gateways for a reason. External gateways offer flexible deployment options including trunk failover and redundancy. Gateways can be deployed anywhere on the network including in different locations. You can add as many trunk lines you need not limited by how many PCI cards fit into a server chassis. Media processing does not load your CPU and media is routed peer-to-peer from the phone directly to the gateway. Gateways are plug & play managed and easy to deploy.
• Interoperability:  sipXecs is a truly SIP standards compliant system using native SIP call control. It is a SIP router that interoperates in a large network and routes calls. Many of the sipXecs developers actively participate in the IETF effort to standardize SIP and have authored or co-authored many of the standards.
• Localization: sipXecs can be easily localized using uploadable language packs. Language packs include voice prompts, full UI translation, local dialplans and region specific call progress tone settings.
• Better voice quality:  sipXecs routes media peer-to-peer and not through the sipXecs server. This has many key advantages among them better voice quality, unlimited number of simultaneous calls, unlimited video calls, works with any codec supported by the end-points, and the PBX is no a single point of faulure which allows load-sharing redundancy.
• Web Services, SOA, and IT integration:  sipXecs offers many interfaces. It is designed to integrate into an advanced IT environment including Web Services based on SOAP for all configuration. sipXecs includes integration with Microsoft AD and Exchange.

We don't release software before it passes a rigorous test process. There is no such thing as bug-free software and testing has to be an integral part of the development process itself. sipXecs is engineered for testability throughout. With about 1 million lines of code that is a necessity and not just nice to have.

• Unit tests: All the sipXecs components have a very high unit test coverage. These are many hundred tests that run fully automatically every time the sipXecs application is compiled. Every new addition to the code comes with a corresponding unit test. Continuous builds excercise all unit tests many times every day making sure that every addition to the code base does not break existing capabilities.
• Automated regression tests:  The sipXecs project team uses a sophisticated mostly automated regression test environment that includes stress and performance testing of all components.
  
• New feature testing: Every new feature added to the code base is accompanied with a new set of test cases. Testing is done on an on-going basis in parallel with development but executed by a separate independent team.
• 6 to 8 months release cycles: Our release process is pretty predictable and it is split into 3 month of new development where new features are moved into the code base and then a 3 to 4 month period of intensive testing prior to release. We also maintain released versions and issue maintenance releases if necessary. This process gives us very high development velocity.