Performance Management (APM) for Cascading / Scalding in Hadoop

Build Better Cascading or Scalding Applications for Hadoop and Continuously Monitor Their Performance

Until now, enterprises deploying business critical Cascading, Scalding and Cascalog applications into production had limited operational visibility because the available tools only monitor cluster performance, not application performance. DRIVEN uniquely provides performance management for Cascading applications to let you manage, monitor and control your data application processes by visualizing the entire data process, end-to-end, and providing a robust set of metrics to surface performance issues that may lead to slowdowns or failures.

“Driven helps us improve the quality, reliability and performance of our applications…”
A
René X. Parra – Principal Architect – HomeAway





Read the Case Study



 

APM for hadoop Cascading
A

Real-Time Operational Visibility

  • See how your applications really behave at runtime
A
  • Real-time monitoring of Cascading, Scalding and Cascalog queries with workflow visualization
A
  • Rich performance metrics provide deep insights into potential performance optimizations
A
  • Easily correlate application performance anomalies with other simultaneous events in the cluster to identify resource constraints and application bottlenecks
A
  • Create and share custom views so everyone is working from the same operational data when troubleshooting problems

Visual Debugging and Proactive Monitoring of Your Cascading Applications

  • Configure alerts and notifications of slowdowns and job failures of critical applications
A
  • Visual debugging of a problematic application processes allows you to quickly narrow down the cause of the issue, i.e. is it the code or resource constraints
A
  • Surface the Stacktrace directly from DRIVEN to pinpoint the root cause of performance degradation, down to the line of code, without wading through dozens of log files
  • Use execution history to understand performance trends and confidence to reliably meet SLA’s
DRIVEN's DAG View
DRIVEN's App View

Rich Operational Analytics for Your Cascading  Jobs

  • Gain insights into your test, pre-production or production Cascading jobs including: average duration, mean time to failure, effective parallelism, and much more
A
  • Powerful search means you can search on almost anything, such as Cascading app name, by Job tracker IDs, by table names, etc. and it’s simple and fast
A
  • A built-in, scalable metadata repository enables you to analyze current application performance and execution behavior compared to historic performance

Manage The Business Of Big Data Operations

  • Integrate DRIVEN alerts and notification into your support/ticketing systems and leverage existing processes
A
  • Filter on operational criteria to allow administrators to easily manage and enforce policies.
A
  • Organize and segment your applications based on any user defined metadata, such as a team, division, organization, or application name.
A
  • Improve general resource management as well as reporting for chargebacks
DRIVEN's Homepage