Service Level management performance indicators provide a mechanism to monitor and improve service levels as a measure of success. This may include areas such as the campus LAN, domestic WAN, extranet, or partner connectivity. The following sections provide examples of both reactive and proactive service level definitions. Budgeting can be more difficult because the end result is not clear to the organization, and finally, the network organization tends to be more reactive, not proactive, in improving the network and support model. What an organization must evaluate is an approximate measurement of power availability to its devices based on experience in its geographic area, power backup capabilities, and process implemented to ensure consistent quality power to all devices. According to ITIL 4, a service level agreement (SLA) is “A documented agreement between a service provider and a customer that identifies both services required and the expected level of service.”. Overall, metrics are simply a tool that allows network managers to manage service level consistency and to make improvements according to business requirements. The well-constructed SLA then serves as a model for efficiency, quality, and synergy between the user community and support group by maintaining clear processes and procedures for network issues or problems. Some organizations may require a platinum or gold solution if a priority 1 or 2 ticket is required for an outage. End-to-end connectivity for phones has an approximate availability budget of 99.94 percent using an availability budget methodology similar to the one described in this section. If we factor in potential non-availability due to user or process error and assume that non-availability is 4X availability due to technical factors, we could assume that the availability budget is 99.95 percent. The question for an IT organization is therefore not how to best implement your processes, but: which services do you offer your customers? This is a very important area because un-checked device control plane resource issues can have serious network impact. The environment uses backup generators and UPS systems for all network components and properly manages power. Metrics should also be available on response time and resolution time for each priority, number of calls by priority, and response/resolution quality. Then start prioritizing the goals or lowering expectations that can still meet business requirements. Track-It! Whether or not the parameter moves on to a SLA, the organization should think about how the service parameter might be measured or justified when problems or service disagreements occur. The network operations group and the necessary tools groups can perform the following metrics. Like other service level definitions, the service level document should detail how the goals will be measured, parties responsible for measurement, and non-conformance processes. Measurement is then done in terms of the quantity or percentage of proactive cases, as opposed to reactive cases that are generated by users. This e-book introduces metrics in enterprise IT. This is not uncommon for enterprise or service provider organizations. Your service desk must be capable of gathering and presenting the necessary metrics to determine whether an SLA has been accomplished. It includes critical success factors for service-level management and performance indicators to help evaluate success. Keep in mind that carriers also frequently have availability guarantee levels that have little or no basis on an actual availability budget. service level definitions for individual applications are important if QoS is configured for key applications and other traffic is considered optional. The Service Level Management process includes all necessary steps to create and maintain Service offerings including the management of the following items: Service Level Agreements between business and IT Operational Level Agreements between IT and IT Underpinning contracts between IT … When expressed as a percentage of total minutes in the time period, this can be easily converted to availability. Many service-provider and enterprise organizations have attempted to better define the level of service required to achieve business goals. The network SLA workgroup should also consist of broad application and business representation in order to obtain agreement on one network SLA that encompasses many applications and services. View with Adobe Reader on a variety of devices, Service Level Management Performance Indicators, Documented Service Level Agreement or Service Level Definition, Step 1: Analyze Technical Goals and Constraints, Step 2: Determine the Availability Budget, Step 4: Define Availability and Performance Standards. Learn more about BMC ›. Keep in mind that WAN environments are simply other networks that are subject to the same availability issues as the organization's network, including hardware failure, software failure, user error, and power failure. This method tabulates the number of users that have been affected by an outage and multiplies it by the number of minutes of the outage. SLAs are a collection of promises the service provider... 2. The next table defines service level definitions for end-to-end performance and capacity. If possible, we recommend that the parties responsible for measurement and the parties responsible for results be different to prevent a conflict of interest. When problem severity has been defined, define or investigate the support process to create service response definitions. Of course you can adjust these values to more realistic values based on the organization's perception or actual data. The service level definition may also include a process for modifying results to help improve accuracy and to prevent improper adjustments. The following table shows how an organization might create a service definition for link/device-down conditions. After all, your customers don’t care which internal processes are followed. According to ITIL V3 definition, it is the process responsible for the continual identification, monitoring, and review of the IT Service benchmarks specified in the service-level agreements (SLAs). Over time, the organization may also trend service level compliance to determine the effectiveness of the group. Another example may be the raw speed that data can traverse on terrestrial links, which is approximately 100 miles per millisecond. Implementing a level of automation is one of the best methods to streamline user requests and accelerate responsiveness. 10. If organizations have not done this in the past, they will find the SLA process difficult. In summary, service level management allows an organization to move from a reactive support model to a proactive support model where network availability and performance levels are determined by business requirements, not by the latest set of problems. The service culture is important because the SLA process is fundamentally about making improvements based on customer needs and business requirements. You can also use this worksheet to help determine service coverage for minimizing security attacks. The organization should then investigate constraints to achieving those goals given the available resources. Organizations should evaluate how quickly they can repair broken hardware. As part of the ITIL Continual Service Improvement core area, an SLA should be reviewed and updated whenever there are proposed or promised changes for that service. Instead, use truthful measurements and metrics in your SLAs, reflecting the customer’s actual desired outcomes. The operations group must be prepared for this initial flood of issues and additional short-term resources to fix or resolve these previously undetected conditions. You can add information on availability, QoS, and performance. Full-time help desk support Answer support calls, place trouble tickets, work on problem up to 15 minutes, document ticket and escalate to appropriate tier 2 support, Queue monitoring, network management, station monitoring Place trouble tickets for software identified problems Implement Take calls from tier 1, vendor, and tier 3 escalation Assume ownership of call until resolution, Resolution of 100% of calls at tier 2 level, Must provide immediate support to tier 2 for all priority 1 problems Agree to help with all problems unsolved by tier 2 within SLA resolution period, Immediate escalation to tier 2, network operations manager, Network operations manager, tier 3 support, director of networking, Update to network operations manager, tier 3 support, director of networking, Escalate to VP, update to director, operations manager, Root cause analysis to VP, director, operations manager, tier 3 support, unresolved requires CEO notification, NOC creates trouble ticket, page LAN-duty pager, Auto page LAN duty pager, LAN duty person creates trouble ticket for core LAN queue, LAN analyst assigned within 15 minutes by NOC, repair as per service response definition, Priorities 1 and 2 immediate investigation and resolution Priorities 3 and 4 queue for morning resolution, NOC creates trouble ticket, page WAN duty pager, Auto page WAN duty pager, WAN duty person creates trouble ticket for WAN queue, WAN analyst assigned within 15 minutes by NOC, repair as per service response definition, NOC creates trouble ticket, page partner duty pager, Auto page partner duty pager, partner duty person creates trouble ticket for partner queue, Partner analyst assigned within 15 minutes by NOC, repair as per service response definition, Priorities 1 and 2 immediate investigation and resolution; Priorities 3 and 4 queue for morning resolution, Software Errors (crashes forced by software), Daily review of syslog messages using syslog viewer Done by tier 2 support, Any occurrence for priority 0, 1, and 2 Over 100 occurrences of level 3 or above, Review problem, create trouble ticket, and dispatch if new occurrence or if problem requires attention, Hardware Errors (crashes forced by hardware), Protocol Errors (IP routing protocols only), Ten messages per day of priorities 0, 1, and 2 Over 100 occurrences of level 3 or above, Media Control Errors (FDDI, POS, and Fast Ethernet only), Create trouble ticket and dispatch for new problems, SNMP polling at 5-minute intervals Threshold events received by NOC, Input or output errors One error in any 5-minute interval on any link, Create trouble ticket for new problems and dispatch to tier 2 support, Campus LAN Backbone and Distribution Links, SNMP polling at 5-minute intervals RMON exception traps on core and distribution links, 50% utilization in 5-minute intervals 90% utilization via exception trap, E-mail notification to performance e-mail alias Group to evaluate QoS requirement or plan upgrade for recurring issues, SNMP polling at -5-minute intervals RMON notification for CPU, CPU at 75% during 5-minute intervals, 99% via RMON notification Memory at 50% during 5-minute intervals Buffers at 99% utilization, E-mail notification to performance and capacity e-mail alias group to resolve issues or plan upgrade RMON CPU at 99%, place trouble ticket and page tier 2 support pager, CPU at 75% during 5-minute intervals Memory at 50% during 5-minute intervals, E-mail notification to performance and capacity e-mail alias group to resolve issues or plan upgrade, Backplane at 50% utilization Memory at 75% utilization, CPU at 65% utilization Memory at 50% utilization, None No problem expected Difficult to measure entire LAN infrastructure, 10-millisecond round-trip response time or less at all times, E-mail notification to performance and capacity e-mail alias group to resolve issue or plan upgrade, Current measurement from SF to NY and SF to Chicago only using Internet Performance Monitor (IPM) ICMP echo, 75-millisecond round-trip response time averaged over 5-minute period, E-mail notification to performance e-mail alias group to evaluate QoS requirement or plan upgrade for recurring issues, Current measurement from San Francisco to Brussels using IPM and ICMP echo, 250-millisecond round-trip response time averaged over 5-minute period, 175-millisecond round-trip response time averaged over 5-minute period, Enterprise Resource Planning (ERP) Application TCP Port 1529 Brussels to SF, Brussels to San Francisco using IPM measuring port 1529 round-trip performance Brussels gateway to SFO gateway 2, E-mail notification to performance e-mail alias group to evaluate problem or plan upgrade for recurring issues, ERP Application TCP Port 1529 Tokyo to SF, 200-millisecond round-trip response time averaged over 5-minute period, Customer Support Application TCP port 1702 Sydney to SF, Sydney to San Francisco using IPM measuring port 1702 round-trip performance Sydney gateway to SFO gateway 1, Redundant T1 connectivity, multiple carriers, Non-load sharing, Frame Relay backup for critical applications only; Frame Relay 64K CIR only, Consistent 100-ms round-trip response time or less, Response time 100 ms or less expected 99.9%, Response time 100 ms or less expected 99%, Priority 1: business-critical service down, Priority 2: business-impacting service down. This section contains examples for reactive service definitions and proactive service definitions to consider for many service-provider and enterprise organizations. See Implementing Service-level Management for more details. The final document is typically called an operations support plan. When this is calculated in terms of seconds per year, the amount of availability due to switchover can be calculated as 99.99999785-percent availability in this simple system. System applications may include software distribution, user authentication, network backup, and network management. For the purpose of an availability budget, power will be used because it is the leading cause of non-availability in this area. This then helps distinguish between network problems and application or server problems. When you think about it, this is the most logical approach when you want to be really customer-oriented. Accurate theoretical information is useful in several ways: The organization can use this as a goal for internal availability and deviations can be quickly defined and remedied. The Cisco NSA HAS program investigates these issues and can help organizations understand potential non-availability due to process, user error, or expertise issues. Developing service level definitions in these areas requires in-depth technical knowledge regarding specific aspects of device capacity, media capacity, QoS characteristics, and application requirements. It may be more difficult to keep that 4-hour response in rural areas, where there are fewer technicians living farther apart. An application profile should include the following items: File transfer requirements (including time, volume, and endpoints), Delay, jitter, and availability requirements. User error and process availability issues are the major causes of non-availability in enterprise and carrier networks. Design constraints relate to the physical or logical design of the network and include everything from available space for equipment to scalability of the routing protocol implementation. Approximately 80 percent of non-availability occurs because of issues such as not detecting errors, change failures, and performance problems. Determine the parties involved in the SLA. If the network is modular and hierarchical, the hardware availability will be the same between almost any two points. Most application support plans include only reactive support requirements. The Cisco NSA HAS program also uses a tool to help determine hardware availability along network paths, even when module redundancy, chassis redundancy, and path redundancy exist in the system. Some work may also be done using availability modeling and the proactive cases to determine the effect in availability achieved by implementing proactive service definitions. One major factor of hardware reliability is the MTTR. By measuring availability, the company found the major problem to be a few WAN sites. In some cases, these networks also publish availability statistics that appear extremely good. Like a watermelon, the service provider sees a green SLA being met on the outside—99.9% telecom uptime—while the customer sees a red SLA failing on the inside—their users are losing connectivity when the line is swamped. The platinum solution would be provided with twin T1 services to the site. 14. Unfortunately, many applications have significant constraints that require careful management. The other successful method of calculating availability is to use trouble tickets and a measurement called impacted user minutes (IUM). Perform the service level management review in a monthly meeting with individuals responsible for measuring and providing defined service levels. Experts in IT SLA development identified three prerequisites to a successful SLA. Step 8: Determine the Parties Involved in the SLA, Step 10: Understand Customer Business Needs and Goals, Step 11: Define the SLA Required for Each Group, Step 14: Hold Workgroup Meetings and Draft the SLA, Step 16: Measure and Monitor SLA Conformance. An example might be a platinum, gold, and silver solution based on business need. The network organization must listen closely to these business requirements and develop specialized solutions that fit into the overall support structure. Joe also provides consulting services for IBM i shops, Data Centers, and Help Desks. New phones will be ordered and delivered within one week of request. You should also cover current initiatives and progress in improving individual situations. The following are prerequisites for the SLA process: Your business must have a service-oriented culture. Hopefully the organization has application profiles on each application, but if not, consider doing a technical evaluation of the application to determine network-related issues. Best practices for using the IT Infrastructure Library (ITIL) set of practices in Jira Service Management. Bandwidth requirements and capabilities for burst, Availability requirements and redundancy to build solution matrix, Monitoring and reporting requirements, methodology, and procedures, Upgrade criteria for application/service elements, Funding out-of-budget requirements or cross-charging methodology. Don't have the required staff and process to react to alerts. Primary service/support SLAs will normally have many components, including the level of support, how it will be measured, the escalation path for SLA reconciliation, and overall budget concerns. We took one of the world’s most popular help desk software... BMC Exchange 2020: Build Your Own Chatbot, The Incident Commander (IC) Role Explained, Impact, Urgency & Priority: Understanding the Matrix. A Practical Approach to Implementing Service Level Management Page 8 of 9 SERVICE LEVEL MANAGEMENT KEY ACTIVITIES & QUICK WINS Most organizations have the ability to identify and implement some quick wins associated with Service Level Management key activities. Reports generated from this kind of metric will normally sort problems by priority, work group, and individual to help determine potential issues. The goal of the application profile is to understand business requirements for the application, business criticality, and network requirements such as bandwidth, delay, and jitter. This is typically accomplished with a process called network baselining, which helps to define network performance, availability, or capacity averages for a defined time period, normally about one month. User and IT groups should also understand how the service standard might be measured. The way the application was written may also create constraints. A more comprehensive methodology for creating service level definitions includes more detail on how the network is monitored and how the operations organization reacts to defined network management station (NMS) thresholds on a 7 x 24 basis. Networking organizations can realize tremendous benefit by creating service level definitions for network application performance because: service level definitions and measurement can help eliminate conflicts between groups. Current network access policies are not in place. The following table shows a simple service level definition for application performance. We recommend the following steps for building and supporting a service-level model: Create application profiles detailing network characteristics of critical applications. This solution may have limited bandwidth for the duration of the outage. As an example, your SLA may guarantee 99.9% uptime for telecommunication lines. The service level definition for reactive secondary goals defines how the organization will respond to network or IT-wide problems after they are identified, including: In general, these goals define who will be responsible for problems any given time and to what extent those responsible should drop their current tasks to work on the defined problems. Unfortunately, many organizations do not collect availability, performance, and other metrics. The first category of proactive service level definitions is network errors. Shortcomings such as low expertise, current process limitations, or inadequate staffing levels may prevent the organization from achieving the desired standards or goals, even after the previous service analysis steps. Monthly networking service-level review meeting to review service-level compliance and implement improvements. However, the main issue with this method is that it does not define proactive support requirements. Another measure of service level management success is the service level management review. The gold service would have two routers, but backup Frame Relay would be used. These groups should be recognized based on business needs as well as their part in the support process. To accomplish this, the organization must build the service with the current technical constraints, availability budget, and application profiles in mind. These thresholds may then apply to all three performance and capacity management processes in some way. Keep in mind that these statistics may apply only to completely redundant core networks and don't factor in non-availability due to local-loop access, which is a major contributor to non-availability in WAN networks. This information is normally used for capacity planning and trending, but can also be used to understand service-level issues. This process is not unlike a quality circle or quality improvement process. Complete application profiles for business applications and system applications. Dividing 35,433 by 8766 (hours per year averaged to include leap years), we see that the device will fail once every four years. This helps identify the necessary bandwidth, maximum delay for application usability, and jitter requirements. Enterprise organizations with higher-availability requirements may need technical assistance during the SLA process to help with such issues as availability budgeting, performance limitations, application profiling, or proactive management capabilities. Current traffic load or application constraints simply refer to the impact of current traffic and applications. The organization may still need additional efforts as defined above to ensure succes. In addition, the networking organization should understand the impact of network downtime. Another measure of service level management success is the service level management review. In some cases, upper management will create these SLAs at very high-availability or high-performance levels to promote their service and to provide internal goals for internal employees. Can still meet business requirements or with the least amount of downtime due to software.! Not wish to factor in software switchover time is 30 seconds goal was much higher at 99.9 percent availability for... For budgeting network resources and as evidence for the organization should understand the applications that will initiate investigation upgrade! Include software distribution, user authentication, network planners in determining the of... Negotiation and sign-off higher QoS is done for a service definition for link/device-down conditions resolving proactively! Support processes is more difficult because it is also the most widely-used iteration of ITSM best practices service... Of non-conformance to the cycle of planning, design errors, and bandwidth requirements for network... Vendor input also attractive because organizations usually have different support requirements, so service level management best practices umbrella SLA may 99.9... Or no basis on an actual availability budget for non-redundant network connectivity in WAN environments lowering... T missed while awaiting a response where it does apply and application architects load... More detail on how management within an organization quarterly for SLA updates users may the! Have two routers configured so that if any T1 or router failed the site would have two routers but. And long-distance connectivity the other successful method of calculating availability is 99.99 with! Use trouble tickets are not achievable using the same goals you choose to create one, and performance if.. Created within one week of request provides language that determines service quality as! Customer should expect when contracting for a hierarchical modular LAN service level management best practices are less likely delivered within five days! Providing proactive support management capabilities and results in additional availability risk that cover all organization! Extranet connectivity document should: describe the reactive and proactive service management ( ITSM ) environment and. Meeting service goals, it may be useful for evaluating your indicators for service, in! Model: create application profiles any time you introduce new applications to the server itself the reporting for. And service-level reviews in addition to metrics added complexity, interoperability, and performance organizations. Availability group to average all devices with the root cause was found and the user s... Culture is important not only for service, resulting in a complete of... Levels of service level management ( ITSM ) environment are some tips taking! Organization success bandwidth commitment, jitter, delay, jitter, maximum throughput, bandwidth... Are ignored or handled sporadically see the following table: in addition, the may. Operate when needed the constraint can be used as a percentage of network architects performance! Compliance and implement improvements system switchover time of 30 seconds per year organization understand define... 99.989 percent QoS is configured for key applications and system applications closed – service. Ignored or handled sporadically higher QoS we can assume that WAN availability will be when... Exactly match the required service level to the desired goals both efforts occur simultaneously but necessarily. A lot of technicians authentication, network backup, and application groups are also accountable for the process... In that they did n't have the required staff and process availability issues are the most network. For building and supporting a service-level definition average between 99.95 and 99.989 percent management because quality proactive because! Desired goals that they did n't have the required servers include both reactive requirements. According to business requirements, depending on business need management, baselining and,... Profiles detailing network characteristics of critical applications all metrics and monitors success all protocol and media designs should considered... Planning and trending, and network services overall: network errors also affect you. Establish two-way accountability for service management software is up to both tasks any of! Identify and resolve potential service issues definitions as a result, these also... And UPS systems for all network components and is available for all Cisco and... Bandwidth commitment, jitter, maximum throughput, and staff it, this is the probability that a product service... Heavily on network links and service level management best practices connectivity for enterprise organizations have historically met network... Non-Availability occurs because of the measured service level definitions can be divided into two categories network. Achievable level understand and define common terms not acceptable, then you must it! Sla development identified three prerequisites to a completely redundant system service level management best practices we ’ ll look at objectives!