AppDynamics 3.0 Interview
Modern application architectures and production environments have changed dramatically in the last 5 years.
Jyoti founded AppDynamics in early 2008 with the vision of defining the next-generation of application performance management (APM) solutions for distributed applications running in cloud, virtual, and physical environments. Before founding AppDynamics, Jyoti led the design and architecture for several products at Wily Technology. Wily was the pioneer and leader in the Java performance diagnostics space, and was acquired by Computer Associates for $375 million. Before Wily, Jyoti worked in senior engineering roles at Datasweep (acquired by Rockwell Automation) and netLens (acquired by Microsoft through FAST/Nextpage), where he built distributed application architectures for enterprise search and data mining. Jyoti received his BS in Computer Science from the Indian Institute of Technology, Delhi. He is the lead inventor on 14 US patent applications in the field of distributed applications management.
JAXenter.com talks Application Performance Management, leak detection and secret-sauce, with founder and CEO of AppDynamics, Jyoti Bansal…….
JAXenter: What makes AppDynamics unique in the
world of Application Performance
Jyoti Bansal: Modern application architectures and production environments have changed dramatically in the last 5 years. Today’s applications are highly-distributed, service-oriented, use open source components and advanced caching strategies, and change frequently because of agile development. Today’s production environments often run across hybrid cloud, virtual and physical data center environments. What this means is that applications are now harder to manage than ever because a) there are more components to manage and b) there is a higher rate of change.
AppDynamics is the only solution built specifically for these dynamic applications. Legacy solutions that require users to set manual instrumentation, set manual thresholds and manually trigger diagnostic sessions quickly develop blind-spots in dynamic environments. With more to manage and a higher rate of change, our focus has been to add intelligence to dynamically discover, instrument and monitor these applications. One of our customers releases 2500 builds per month and another customer fluctuates their cloud capacity by 100’s of nodes per day. Our ability to dynamically adapt to application & infrastructure change makes us quite unique.
JAXenter: How does AppDynamics 3.0 build on the
product’s leak detection
Jyoti Bansal: As companies re-architect their mission-critical applications to leverage caching strategies and improve performance, they realize they need new solutions to better monitor and troubleshoot memory leaks and trash. What they find is that legacy solutions, like Java profilers, are not a good fit for production environments. For applications in production, AppDynamics 3.0 provides real-time monitoring of Java heap and garbage collection to automatically detect memory leaks as they happen. What makes us unique is our ability to diagnose the root cause by analyzing the content of the memory structure and determining which code paths or transactions are causing memory to grow.
JAXenter: What is ‘Deep-on-Demand Diagnostics’?
Jyoti Bansal: Deep-on-Demand Diagnostics is our patent-pending algorithms that are core to our secret-sauce. There are three underlying algorithms designed to provide the following customer benefits:
1) In-flight Response Analysis: Whenever a problematic session
AppDynamics will collect diagnostic data.
2) Performance Pattern Analysis: Accurately distinguishes
from consistent patterns of poor performance.
3) Call Stack Analysis: Intelligently collects code-level
less than 2% overhead.
JAXenter: Who are you targeting, with AppDynamics 3.0?
Jyoti Bansal: AppDynamics can help any enterprise that has mission-critical or revenue-critical Java applications running in their organization or in the cloud. Some of our customers include companies like Priceline, TiVo, NASA, Taleo, etc. With AppDynamics 3.0, we become the standard bearer for performance management in the cloud. We’ve worked very closely with Netflix over the past 7 months as they rolled out their movie streaming applications in cloud environments and are well equipped to handle the unique performance challenges of any large-scale cloud application deployments.