Capacity Optimizer automatically tunes and optimizes cluster resources
Automatically Improves Big Data Cluster Performance without Manual Tuning
Continuously tune and optimize your big data cluster resources. Recapture wasted capacity so you can run more applications and get the most out of your infrastructure investment.
Improve Big Data Cluster Throughput By Up to 50%
By monitoring the entire infrastructure in real time and leveraging active resource management, Capacity Optimizer identifies where more work can be done, and adds tasks to servers with available resources.
On a typical cluster, Capacity Optimizer makes thousands of decisions per second on how to best optimize your big data clusters, increasing typical enterprise throughput by up to 50 percent. Even the most experienced operator can’t make manual configuration changes with the precision and speed of Capacity Optimizer.
Doesn’t YARN Scheduler Manage Resources?
YARN (“Yet Another Resource Negotiator”) scheduler leverages the resource management capabilities of MapReduce, coordinating consumption and usage reservations to make allocations. Limited by its conservative assumptions about memory usage, YARN under-provisions resources. In addition, YARN does not monitor containers once they start running or adjust in real time based on actual usage.
Pepperdata Capacity Optimizer overcomes YARN scheduler’s limitations and maximizes resource utilization by monitoring actual per-task hardware usage in real time. Capacity Optimizer dynamically makes adjustments at the process level to eliminate inefficiencies and bottlenecks, and maximize resource usage.
- Ensure big data cluster stability and efficiency.
- Avoid overspending on hardware.
- Reduce time spent on capacity planning.
- Run more jobs on existing infrastructure.
- Run more jobs faster.
- Access additional cluster capacity.
- Spend less time in backlog queues.
- Report on capacity trends
- Get accurate chargeback reporting
- Increase productivity