Big Data systems by their very nature are complex entities. This is because everything about them is super-sized: the data volume, the number of machines, and the velocity of the data. As a result, building production Big Data systems is difficult. It requires developing new intuition about both development and operations.  And, it requires the realization that performance is paramount.

Over the past several years, Pepperdata has been helping our customers integrate a deep understanding of performance into their operations teams. While this has proved indispensable for running their Big Data production systems, customers have asked us to provide feedback to developers to address issues as early as possible. Developers make choices in algorithms and code that have profound performance and operational implications. Unfortunately, few developers receive simple, actionable feedback for preventing performance issues before they launch an application into production. This is precisely the gap that Pepperdata Application Profiler fills.

Application Profiler represents a broadening of the Pepperdata Product Suite. It provides simple, actionable steps that developers can take to improve or avoid performance problems in production systems. The core of the new product is the open source project Dr. Elephant initiated by LinkedIn.  Pepperdata is providing a hosted version of Dr. Elephant integrated with our Cluster Analyzer. The integration is powerful in that it provides actual run-time context for developers within which to understand performance improvement recommendations.

The use of Dr. Elephant also represents a strategic choice by Pepperdata to embrace the open source community. It is our intent to vigorously contribute changes as well as customer input back into Dr. Elephant. We are delighted by the warm welcome we have received from LinkedIn and look forward to being an active participant.

Strategically, we see ourselves being at the forefront of delivering products that are indispensable for operating Big Data systems in production. Our intent is to address the entire DevOps cycle offering solutions that work well on-prem and in the cloud. Our future will leverage the enormous data volume of metrics that we have collected (over 20 trillion data points), along with best practices and machine learning to enable operational systems that are simultaneously reliable, scalable and performant.

Keep an eye out for a steady stream of exciting new announcements from Pepperdata.