Cloud Service Providers need to proactively manage service quality and profitability
Cloud services are a new class of computing services noted for their dynamically scalable, virtual computing resource offered as a service over the Internet. Cloud service providers promise to hide the complexity and cost of owning and operating high performance, global IP and computing infrastructures and rely on pre-defined SLAs to match customer demand with service features and price.
Since both the cloud service provider and the hosted VoIP service provider deliver real time services over a shared IP infrastructure they encounter a broad array of operational and customer satisfaction challenges including packet prioritisation, bandwidth limitations on service quality and other challenges consistent with real time service delivery over IP. Leading cloud service providers need an advanced service performance management solution to assure continuous integrity of their service design. Cloud service provider support staff needs to adjust operating parameters that make service performance predictable and therefore economically viable.
Psytechnics Experience Manager enables shared, continuous service monitoring and proactive troubleshooting
| • | Pro-active, real time service level monitoring and management based on predictive impacts of network problems on actual experience |
| • | Pin point diagnoses allow operations staff to take the right corrective action, first time |
| • | Built-in support for multiple customers from the same operations platform, reducing costs, and improving efficiency |
Psytechnics makes the difference for cloud service providers
At the network level, both cloud services and real time person-to-person communications services have many similar requirements – minimal jitter, delay and packet discards – that lead to high quality experiences. The presence of other traffic, integration practices of many networks and the unique configuration and operating policies of many discrete network elements require the adoption of continuous monitoring technologies that provide information about the operational state of the service, as experienced by the end user.Psytechnics is a recognized leader in technologies for the management of real time IP applications and services. Real time services have far more complex delivery attributes, particularly in shared network environments and in human-participating communications than typical store-and-forward applications and computer-computer communications. Many implementations have failed to deliver customer satisfaction, or failed to deliver the expected profitability, because of underestimated support requirements. These shortfalls and additional unforeseen overheads can not only create customer satisfaction issues but worse still, often trigger SLA penalties or reduce yields on pricing segmentation strategies.
The cloud service provider must be armed with information about the end to end service performance even when the service traverses an intermediate network infrastructure . Their operations engineers must be in a position to confidently articulate to third party service providers and even to their cloud service customers those service-affecting issues in their specific domains, and to confirm that changes made have corrected the situation. Real time information about service state and network performance helps to focus all service participants onto the policies, practices and faults that detract from optimal service and therefore profitability performance. To do any less is to deny the cloud service provider and their service customers the best this nascent industry has to offer.
The right architecture assures that the right information is available
Cloud service providers are a statistical business. Like a telephone company, they over-subscribe their fixed and extensive compute resources among many customers. Like an applications manager at a financial institution, the cloud service provider operations engineers continuously monitor meta-data (information about the data) and reactive quickly to remote and sometimes unconnected events that end up affecting the quality of the service.Cloud service providers must scale their operations support implementation as well as their computing service. Support engineers must be capable of the simultaneous supervision of thousands of customer sessions, so they can focus their skills on the high priority and high cost issues.
The Psytechnics Experience Manager architecture provides for the deployment of appliance based passive Experience Engines at key traffic points. These engines analyze real time calls and funnel key service metadata to the centralized Experience Server, which stores the relevant data for historical analysis, reporting and archiving. Through the use of APIs the flow of relevant state and operational information is presented into umbrella systems and screens. Through a web-GUI, the Experience Manager presents the ongoing state of the service experience overall, and as necessary on a session-by-session basis. Oftentimes, through the tuning of Experience Manager alarm settings, support engineers are able to be alerted to impending events and take corrective action on network problems, before the customer experience is degraded.
By collecting and analyzing real time service data an accurate picture of aggregate and customer-specific service levels and therefore peak operating performance can be formed and maintained. Just as importantly, should problems occur, the Psytechnics solution can accurately pin-point what the problem is in order to effect a rapid resolution.
Customer support requests are matched to a monitored problem, the extent to which other customers are affected is determined, and the root cause is identified. Cloud service providers often do not have the time or the skills to attempt problem isolation with real time communications services with typical computing supervision tools.
They need the ability to view the network service as a whole – including computing operations and network operations – especially if they don’t control the end-to-end service.
Many times, determining where the problem isn’t can be just as important as determining where it is. This can happen when the responsibility for fault resolution may lie with the customer, a network service provider or some external third party. Identifying the right resource, or provider, can make an enormous difference to the resolution of problems in a timely and contractually committed manner. Psytechnics makes that difference for our customers every moment of every day, for over a million end users under management.