Get a Comprehensive View of Your Applications, Infrastructure, and Resource Utilization
Pepperdata Platform Spotlight gives you an in-depth view of your applications, infrastructure, and resource utilization in any cloud and on-premises. It also provides tuning recommendations and custom alerts, so you can rapidly make resource decisions and diagnose performance issues. Pepperdata cluster performance monitoring includes real-time and historical information covering system demand, abusive users, and wasteful applications. Platform Spotlight allows you to:
- Monitor supported technologies and frameworks, including Kafka, Impala, HBase, Hive, MapReduce, Kubernetes, Tez, Spark, Amazon EMR, and Google Dataproc.
- Identify the impact that applications have on cluster-wide resources such as CPU, disk, memory, network, NameNode, and HDFS.
- Improve troubleshooting by identifying application slowdowns caused by disk, network, or node failures/congestion, and quantify the impact the cluster has on applications.
Optimize Workloads with Recommendations
Big data operators need to configure and size critical resources running on big data environments, but it’s hard to get an accurate view without access to the right data. It’s even harder when operating in complex cloud settings. To make this easier, Pepperdata provides actionable reporting, insights, and recommendations to rightsize containers, queues, and other resources.
Platform Spotlight does this by continuously collecting extensive unique metrics and data about your hosts, queues, users, applications, and other relevant resources, providing you with insights to quickly diagnose performance issues and make resource decisions. Platform Spotlight allows you to:
- Size big data queues appropriately based on criticality.
- Get self-service recommendations on container sizes and heap reservations.
- Identify growth trends and accurately forecast resource needs.
Observability: What it is and Why it Matters
Observability is an extremely popular topic these days. What’s driving this interest? Why is observability needed? What is the difference between observability and monitoring? Read this blog post to learn the answers and more.
Store Your Data More Efficiently with HDFS Data Temperature Reporting
You probably don’t know the exact age of every file you have stored in HDFS, or which files should be archived or moved to SSDs based on how frequently (or infrequently) they are accessed. And not knowing data temperature can be costly. With Platform Spotlight, you can generate a summary of policy definitions and rolled-up data on the state of HDFS, including the total number of files and where they fall within the policy spectrum. Included is a cold file breakdown that details each policy’s coverage and each file’s state relative to the governing policies. This breakdown includes the exact file names and sizes for each temperature. With the Pepperdata Data Temperature report, you can:
- Optimize your data retention and reduce your storage spend.
- Get a comprehensive view of how much movement is required to adhere to policy.
- Get the age and size of your HDFS data files.
- See exact file names for each temperature.
- See which files don’t match their current policy based on access times.
Create and Receive Alerts on Platform Health and Performance
Platform Spotlight enables you to create and receive alerts about events that degrade performance. You can:
- Quickly perform root cause analysis (RCA) of performance issues and operational inefficiencies.
- Use data on activity such as duration and amount of data processed to alert on rogue applications and identify which applications will miss their SLAs.
- Achieve optimal application and cluster performance on multi-tenant systems.
Monitor Big Data Application Performance and Run More Efficiently on Amazon EMR
Automatically optimize your big data infrastructure and deliver superior application performance in the cloud with Pepperdata for Amazon EMR. Platform Spotlight is available on AWS Marketplace as part of the Pepperdata product suite. In addition to optimizing application performance, get full-stack observability, automated tuning with managed autoscaling, and real-time insights across all of your EMR instances—all in one place. Pepperdata for Amazon EMR allows you to:
- Get full-stack observability, automated tuning, and job-specific recommendations for Spark and MapReduce.
- Use managed autoscaling to automatically optimize node performance and prevent waste by applications.
- Customize alerts to quickly understand and troubleshoot application and infrastructure issues.
Magnite Improves Performance and Streamlines Automated Advertising Solution
Magnite knew they could better manage their clusters, but lacked the granular insight needed to make it happen. Pepperdata Platform Spotlight gave them the granular visibility necessary to quickly pinpoint, troubleshoot, and resolve problems in their cluster.
Benefits for Your Team
Operation Teams
- Understand why jobs are running slowly.
- Compare job runs over time.
- Understand the performance of each job phase.
Application Teams
- Automatically identify bottlenecks and alert on duration, failure conditions, and resource usage.
- Highlight applications that need attention.
Business Teams
- Report on capacity trends.
- Get accurate chargeback reporting.
- Increase productivity.