Event correlation provides a powerful lever to analyze the events data to better manage the enterprise IT systems. While many event correlation algorithms are present in literature, applying them in practice presents many real-world challenges. In this paper, we present our experience with the challenges, the workarounds, and the opportunities in using event correlations in enterprise IT systems. We propose a domain-aware way to select the right scope of events for correlation, with self-tuning parameters and recommending right correlation signatures based on the use-case. We present the effectiveness of our ideas with a real-world case-study.

Today's enterprises heavily rely on their batch systems to ensure smooth business operation. Any delay in these processes has direct impact on sales, revenue, customer experience, and brand image. In this paper, we present an approach to predict these batch processes and generate ahead-of-time notifications of potential batch-induced outages. We present several case-studies to demonstrate the effectiveness of the proposed solution in various real-world scenarios.

Today's IT systems heavily rely on IT support for smoother and faster operations. Any application or infrastructure issue is reported by tools or end-users in the form of a trouble ticket. The information about actual issue is hidden inside these ticket descriptions and is provided in different ways. In this paper, we address the problem of extracting the issues from these ticket descriptions. We break the problem into different sub-problems each presenting different challenges and hence requiring different solutions! We present our experience of applying theory into practice by presenting a real-world case-study.

The management of power consumption in datacenters has become an important problem. This needs a systematic evaluation of the as-is scenario to identify potential areas for improvement and quantify the impact of any strategy. We present a measurement study of a production datacenter from a joint perspective of power and performance at the individual server level. Our observations help correlate power consumption of production servers with their activity, and identify easily implementable improvements. We find that production servers are underutilized from an activity perspective; are overrated from a power perspective; execute temporally similar workloads over a granularity of weeks; do not idle efficiently; and have power consumptions that are well tracked by their CPU utilizations. Our measurements suggest the following steps for improvement: staggering periodic activities on servers; enabling deeper sleep states; and provisioning based on measurement.