VoIP Readiness in the Cloud

VoIP / Video Readiness in the Cloud

The next three years will see the biggest equipment refresh in the history of telecommunications as companies of all sizes begin the migration to IP networks. Replacing old style copper circuits and legacy phone systems with advanced IP features such as Unified Communications is a strategic move that lets you provide better service, better collaboration, better decision making and accelerates the pace of your business.

Is your network able to support the real time impact of voice over IP, high definition video conferencing, real time presence status and mobility solutions? Specialized real time VoIP/Video performance management is an essential requirement from pre-deployment design to planning to implementation to continuous 24x7x365 monitoring.
 

A clean data path is critical to providing a high quality real time VoIP and Video user experience.

 

Characteristics of UC&C

Unified communications and collaboration combine performance sensitive applications such as telephony, unified messaging, mobile communications, high definition video conferencing and real time presence onto a common IP network. Real time communications applications are heavily impacted by bandwidth availability, latency, packet loss and jitter. Even minor performance issues can degrade the user experience. However, mean opinion scores (MOS), latency and packet loss measurements are just one component in understanding what makes a high quality experience for users. Active network monitoring that continuously analyzes each hop along the delivery path is just as important.  
 

4 Styles of Network Readiness Testing

Understanding What You Get

1)     Physical site survey is a manual inventory and bandwidth estimation. Either your network equipment is properly configured, and you have enough bandwidth to support your maximum number of concurrent calls, or you don’t.
2)     Feasibility study is a one time snapshot of your network indicating red light / green light health of the devices and simulated VoIP stress testing that generates synthetic VoIP packets and attempts to overload the network
3)     True voice testing generates real voice calls (RTP and RTCP SIP packets) without deploying IP phones or any other VoIP gear. Voice tests have all the payload and characteristics of real voice calls and are the most accurate way to stress the network by simulating real voice traffic on a path via Layer 4 protocols that put greater loads on the network.
4)     Post deployment, on demand testing is the ability to rerun readiness simulations whenever you deploy new network resources - versus a one time snapshot in time.
 

2 Distinctions with a Path Centric Approach

1)     Abacus Path Assure is a non-invasive path centric approach that tests the network end to end along the path used by the application. A device centric approach looks at the behavior of the device (is it healthy, is the CPU OK, is the memory OK). If the quality issue is upstream of the WAN router and the device’s behavior checks out (no dropped or out of order packets) a device centric approach would not provide alerts or diagnostic information for further troubleshooting. Path Assure provides a readiness value that is an aggregate assessment of all factors across each hop on a network path and provides full mesh end to end visibility across remote sites, distributed users, cloud and internet service providers.
2)     Abacus Path Assure is a lightweight active IP end to end continuous performance monitoring and remote assessment tool. It is not a one time snapshot of your network traffic or a timed stress test that lasts 1-2 weeks on specific sites. Abacus Path Assure lets you run reassessments on your entire network any time prior to deploying new applications, or adding new sites, or resources to your network --- all year long.
 

Implementation

Micro or 1U rack appliances plug into any open LAN port on the network, require zero administration and self register back to the cloud. No permissions, login credentials or authorizations for any of the devices or end points to be monitored need to be collected. No configuration of spanned ports, mirroring, port forwarding or access to secure areas is needed. Path Assure automatically identifies LAN/WAN targets to be monitored.
 
Path Assure can be deployed as a public or private cloud service with no need to purchase, provision, maintain or upgrade any software. Each Linux based micro or rack-appliance can monitor hundreds of network paths in real time.
 

Eureka Questions

1)     Are you rolling out a performance sensitive application such as VoIP/Video?
2)     Any blind spots within your domains or your carriers or anything in the middle?
3)     What applications are consuming what amount of bandwidth?
4)     Who streamlines Netflix or Pandora during the network’s busy hours?
5)     Why do I get 30 MEG when the path is provisioned for 100 MEG?
6)     What is causing the excessive packet loss on VoIP calls in the branch office?
7)     How can I quickly isolate the domain and determine who “owns” the problem?
8)     How can I troubleshoot out-of-order packets that are upstream from the carrier’s WAN router (need path-centric approach versus device-centric approach).
9)     We have persistent call quality issues on Tuesday mornings. Can I replay the network to see who, what, when apps were running going back several Tuesdays?
10)Can I analyze just the hops on the Boston to New York WAN link?
11)Why do rogue IP addresses constantly appear from the conference rooms?
12)Who is responsible for spikes in traffic during off hours?
13)How do I manage the sky rocketing use of Skype Video and Apple Facetime?
 

Path Assure Features - Monitor, Measure, Troubleshoot and Report

1)     Baseline and pre-deployment assessment to verify network readiness for new services. A snapshot in time approach is not sufficient as network utilization and application performance will change over time
2)     Quickly auto discover network devices and pre-assess how well a network can handle new or expanded applications such as real time unified communications
3)     Continuously monitor 7x24x365 performance of on premise, site to site or cloud based services with proactive notification and real time end to end measurements to understand what is happening now
4)     Optimize and troubleshoot with proactive notification to focus resources exactly where needed and keep the network working smoothly
5)     Custom reports validate and quantify network problems - did the changes work?
6)     Verify the enforcement of SLA agreements end to end
7)     Troubleshoot unforeseen problems such as serious connectivity issues, web and cloud services response time, latency on certain network segments that persist despite investment in expensive new technologies
8)     Diagnose and remediate up to 60 network impairments
9)     Storage in the cloud for one full year
10)Visibility into external sources like Salesforce and Google Apps, as well as from workstations out to hosted cloud-based services
11)Works over any IP network - LAN, WAN, WiFi or satellite and you can measure the true network capacity across 3rd party service providers and cloud servers
12)Performs tests every 60 seconds with a very low touch (20 packets per minute) so your applications won’t even know Path Assure is standing guard
13)Remote network and application performance across all data center, cloud, remote office and mobile environments
14)Active bandwidth monitoring, application performance monitoring, remote flow generation and analysis, remote packet capture
15)Expand remote performance management beyond Layer 3 with cloud services that test the capability and performance of Layer 4 traffic on the network
16)No installation, software patches, hardware upgrades
 

Technical Overview

1)     High precision, micro packet bursts are dispersed across the network
2)     Device specific packets are sized, spaced and timed down to nano-seconds
3)     Packet dispersion goes beyond pulling data from device logs
4)     Device-specific packets are sized, spaced and timed to the nano second
5)     Create a specific fingerprint based on 60 distinct network impairments
6)     Complete unbroken view of the network path with full asymmetric uplink and downlink monitoring of performance over WAN/LAN down to each end point
7)     Full mesh view – each site can see each hop on the edge, and the LAN/WAN, and in between sites, and back and forth (if configured for full mesh on every site)
8)     Comprehensive reporting and network playback
 

Conclusion

Studies show that up to 40% of the time spent on a premise based network management tool involves maintaining and managing the system instead of focusing on the network and application management issues that the tool reveals.