<img src="//bat.bing.com/action/0?ti=5739181&amp;Ver=2" height="0" width="0" style="display:none; visibility: hidden;">
Sophie AIOps
Case Study


Automating The Root-Cause Analysis of IT Incidents with Sophie AIOps

 


log-me-in-1

 The Company

Based in Boston, LogMeIn is a provider of SaaS and cloud-based remote connectivity services for IT management, customer engagement, and collaboration. The company's products give users and administrators access to remote computers. LogMeIn has an annual revenue of $1.024B (2017) and nearly 3,000 employees.

The Challenge

One of LogMeIn's product is a complex and sophisticated AI chatbot solution that encounters many unpredictable issues, which directly affect customer experience. The company was reactive toward problems, manually searching logs to uncover root-cause and fix issues to then manually correlate them among vast amounts of log data. Because problems were very difficult to predict, the team was left with significant blind spots even after configuring alerts and creating dashboards. They also saw too much noise from false alerts that were not significant problems. Finally, because LogMeIn uses a DevOps model, time spent troubleshooting problems was a direct trade-off for engineering efforts to advance the product.

The Solution

Because Sophie (Loom’s AIOps) does not need to be configured to find certain problems and instead measures every data point from the logs vs. a behavioral baseline, she is able to uncover significant problems before customers are affected. Sophie also provides automatic correlation between multi-layered issues so the team immediately sees the root-cause of problems; directly improving customer satisfaction and product reliability. 

The Implementation

LogMeIn now streams its logs from production and development servers directly into Sophie in real-time. Logs are parsed automatically via Sophie’s pattern recognition and lexical detection capabilities, including key custom applications that are proprietary to LogMeIn. Sophie then automatically uncovers issues by finding anomalies vs. behavioral baselines, and alerts the right parties to trigger the proper workflows from the team. During daily 15-minute feedback sessions over 2 weeks, Sophie’s alerts were tuned to the proper volume and priority so the team could focus on more important issues.


Results

Correlations

Instead of searching different sources of data between applications, infrastructure, and databases, LogMeIn is able to visualize correlations clearly and immediately with Sophie to discover the root-cause of problems. The LogMeIn team catches several significant multi-layered issues a week, which are potentially customer-impacting and are detected proactively by Sophie.

Noise Reduction 

Within two weeks of 15 minute daily review sessions, Sophie’s alerts were optimized for the team's most important issues and noise was reduced by 75%.

Productivity

Sophie’s “Tribal Knowledge Base” allows users to add insights for any incident. When the same or similar issues re-occur, the insight that was previously added is automatically propagated within the new incident, helping LogMeIn reduce MTTR (mean-time-to- resolution). This saves time and lightens workload for development, DevOps and IT teams.


Why Loom 

  • If you are using a log management tool, but it is either not performing or requires too much knowledge and effort to extract value, Loom's AIOps provides a great option to either augment or replace an existing tool.
  • If you are looking for a proactive tool to detect unknown issues before they impact customers, Loom offers cutting-edge AI log monitoring and analysis.

 

About Loom Systems

Loom Systems’ patent-pending AIOps solution predicts and prevents problems in businesses undergoing a digital transformation or cloud migration. Loom is the only AIOps solution to predict IT issues before they impact customers, and enriches them with insights and resolutions in plain English. This not only keeps operations running smoothly and improves business productivity, but also alleviates the tedium of reading logs and frees up time for operations to concentrate on other important IT matters.

 

See What Loom Can Do for You

Live Demo Get Started