Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Application SLA - the missing part of complete SLA management

How can operators ensure that the proper quality of so many complex services is delivered? Has the software for network and service monitoring enough functionality to provide the right information? Fortunately, OSS systems have evolved, and they currently contain functionalities allowing the
operator to build comprehensive service management platforms. Today, operators cannot even think about delivering modern services of a high quality without providing a SLA. This means that service assurance with SLAs becomes the most critical aspect of modern OSS solutions. Additionally, since most modern services are built based on a number of applications, delivering the services over network, the applications are becoming the core of the service models.

  • Login to see the comments

Application SLA - the missing part of complete SLA management

  1. 1. Application SLA The missing part of complete SLA management?
  2. 2. How can you be sure that the with the number of services delivered they maintain the right quality ? Has the software for network and service monitoring enough functionality to provide the right information? Fortunately, OSS systems currently contain functionalities allowing the operator to build comprehensive service management platforms. Today, operators cannot even think about delivering modern services of a high quality without providing a n SLA – servic e assurance with SLA becomes the most critical aspect of modern OSS solutions. Additionally, since most modern services are built based on a number of applications delivering the services over network, the applications are becoming the core of the service models.
  3. 3. Service Assurance solutions <ul><li>Modern Service Assurance solutions allow for monitoring the service by providing the following key functionalities: </li></ul><ul><ul><li>Service state propagation based on network layer monitoring and service models with detailed information about service-resource relations </li></ul></ul><ul><ul><li>Service Quality Management with statistics gathered from various sources, including the network, probing systems, applications, and supporting systems ; i t also generates service affecting events based on thresholds trespassing </li></ul></ul>These solutions are capable of joining the data in order to provide comprehensive information about the state of the service. This is a must-have solution when the service provider wants to implement the SLA and start including the applications within it.
  4. 4. Service model – dynamic image <ul><li>When a service model includes information on applications, it is important that the model is up-to-date </li></ul><ul><li>The network environment and the applications evolve, so the minimum you can do is to store information about changes that occured </li></ul><ul><li>The ideal solution includes strong integration between Service Inventory, Network Inventory and the monitoring solution; it includes the processes built using ITIL Change Management, that tracks all changes made in resources used by the service </li></ul>These processes should automatically update the service model and trigger the refresh of information in the Service Assurance tool
  5. 5. Service models based on SID
  6. 6. Integration of applications = challenge <ul><li>Challenge: it is possible to receive two events for the same problem </li></ul><ul><ul><li>One originating from the device </li></ul></ul><ul><ul><li>And one originating from the application </li></ul></ul><ul><li>There are two solutions possible: </li></ul><ul><ul><li>A p roper service model, allowing for correlation of these two events and propagating the alarm only if it affects both components of the service </li></ul></ul><ul><ul><li>Correlation rules, which can narrow down the number of events and can enrich and qualify just one event </li></ul></ul>
  7. 7. Monitoring of applications <ul><li>Applications can be monitored in two ways: </li></ul><ul><ul><li>Through the integration layer with dedicated interfaces </li></ul></ul><ul><ul><li>Through IT application management </li></ul></ul>This way you can obtain information on the state of application, performance, disk usage, etc. These detailes can then be included in KPI parameters
  8. 8. SQM & SLA Management <ul><li>When the service model is implemented there are only two more things missing: </li></ul><ul><ul><li>Service Level Management (SLM) </li></ul></ul><ul><ul><li>Service Quality Management (SQM) – collects network and application metrics to calculate KPI and KQI </li></ul></ul><ul><li>This way we have a complete picture of a service: </li></ul><ul><ul><li>Service model as documentation of used components, resources, configurations, and most importantly in current complex services – used applications </li></ul></ul><ul><ul><li>Service impact events providing almost real-time status of the service </li></ul></ul><ul><ul><li>Service KPI and KQI providing the measurements for service quality </li></ul></ul>By including the applications in the model, the overall service information depends on the state and performance of the applications. T his data exported to the SLM solution, can be used by a number of different SLA attributes such as availability, bandwidth or quality .
  9. 9. Application event priorities based on SLA
  10. 10. Customer information prioritizes events <ul><li>Every CSPs tracks the signed SLA, maintains information about the priority of each service and as a result is aware of the potential financial loss caused by a service failure </li></ul><ul><li>The above data can be used to prioritize a given service </li></ul><ul><li>If applications are part of the model, their impact can also be prioritized, so that events caused by them are resolved in an appropriate order </li></ul><ul><li>Application prioritizing can also be used by other users in other activities like enhancement planning, upgrades and other maintenance activities </li></ul>
  11. 11. More and more services depend heavily on applications, like content servers, LDAP servers or even the OS installed on their equipment. These applications often generate more errors than the hardware used in the network. The end-customers only care if the service is working properly and they will change the provider if it doesn’t. Increasing amounts of CSPs use guarantees of quality and availability of the service as a differentiator of their offerings. E very day brings new service requirements and demands - only tools allow ing for flexible service modeling and including applications in management can deal with the complexity of the service offer .

×