Monitoring for Service-level Assurance
Why Monitor an SLA?
Your customers depend on a reliable, always-on presence on the Internet. If your company does not deliver this to them, you lose more than their presence on the Internet. Service outages can also damage your brand and standing in the competitive landscape of monitoring.
Service-Level Agreements (SLAs) alone cannot prevent service outages without some level of monitoring. With monitoring in place, you can improve services and enforce incoming and outgoing SLAs for all devices and customers that depend on you.
The need for monitoring of services increases each day. There are more mobile users, fewer traditional desktops, and the proliferation of millions of Internet of Things (IoT) to track.
All of these devices rely on Application Programming Interfaces (APIs), which shuttle the information between end-users and business applications or services. APIs form the critical core of mobile applications and many web or IoT services. So, monitoring API traffic is vital to maintaining the performance of the increasingly complex web of global applications.
In short, SLAs must cover these now-critical APIs as well; no longer can we confine SLAs to just web service applications.
Due to their interconnected nature, these outages have cascading effects on associated partners, connected websites, and the overall business.
On January 26th, 2015, Facebook’s outage was one such example of needing a comprehensive SLA monitoring program. It carries a powerful message: the need for third-party SLA tracking. There may have been multiple reasons for the outage, but an appropriate SLA should have been in place to protect the interest of all involved parties. The outage had a significant impact on Facebook’s partner sites, including Instagram and Vimeo. It also affected the integration side. Amazon and Dropbox underwent similarly high-profile outages in 2013, with a significant impact on business and revenue.
So, an SLA monitoring program will need to be designed to monitor all facets of your services; your own application services that you provide internally and externally, third-party vendors that your business depends on, and the customer experience that your customers deserve–not a small task. How does a properly designed SLA monitoring program protect your brand and your business?
What is SLA Monitoring?
SLAs are the most critical component of any vendor contract. How do we measure and enforce these SLAs? Using an SLA monitoring tool allows you to create an SLA report informing all parties if the SLA has been met.
An SLA describes all the vendor’s expectations and responsibilities, including the scope and quality of the work. It also clearly defines the penalties — or any acceptable alternative solutions-if the work falls short of these requirements. Think of an SLA as a watchdog for any third-party services. And think of SLA Monitoring as a feedback loop to all parties that keeps them aware of their current SLA attainment.
SLAs are made up of service-level objectives (SLOs) which define all relevant metrics (service-level indicators – SLIs) and their values that you expect to see over time.
Selecting the proper SLI’s to measure is the first step because these aggregated metrics will inform you if you can meet the legal obligations that you make in the SLA. Typically, an SLI will cover performance (time/speed) and reliability (uptime/availability) but can be any metric (e.g., size, bandwidth) that is at the core of the service.
Once you decide on an SLI, you can measure it over time using a script running on an agent for the location and scenario you need compliance. This period measured for this SLI will be the basis for one of your Service-Level Objectives (SLOs, e.g., 99.99% monthly uptime). Your measured SLOs will also provide essential information about whether they will collectively meet the SLA they comprise.
In short, you validate critical SLOs by monitoring SLIs overtime for what your business requires. SLO’s will then cumulatively become the independent source of the SLA’s business-level measurements.
So, the quality of SLA compliance is dependent on how it is measured. For example, to determine service availability in New York set up an hourly SLA tracker for New York’s specific uptime and availability for a service that you own or need compliance with. But New York may be just one location that you need to monitor for comprehensive SLA coverage, so ensure that you measure everywhere your customers are. Service-level dashboards can list the values for each SLA parameter, as agreed between parties so that trends and problems become visible at a glance.
SLAs are contractual obligations and exist to protect all parties involved in the agreement, so protect your legal obligations by ensuring the SLOs and third-party SLAs you’re monitoring are in your contract.
Using Performance Monitoring to Validate SLAs
Your mission-critical business applications need SLA performance and availability monitoring as part of a complete service-level assurance program.
Track your SLA with either our SLA report or our SLA widget, visible on our reporting dashboard, allowing you to quickly monitor and validate your SLA parameters. With all the critical user journeys being covered and checked regularly from different locations, browsers, or API scripts, you can quickly see if your SLAs are in order.
Each successful and failed SLA monitoring check affects the overall SLA percentages. Apica makes it easy for users to see when a check deviates from SLA parameters with only a glance.
Set up your service level reporting to inform all stakeholders of their areas of concern so that the team can help resolve issues.
- Our service-level monitoring platform is highly flexible. Apica can provide service-level assurance on critical transactions for any device, authentication, service, application, or scale.
- Apica provides initial testing and ongoing checks for proactive incident management. Avoid getting trapped after a service incident happens. Use SLA reporting to enforce compliance, retain customers and revenue.