Pepperdata

Client:

Enterprise Software Finteeh

Role:

Lead UX / UI Partner

Tools Used:

Sketch, Axure, Zeplin, Photoshop, Google Docs

Year:

2018-2019

Overview

Pepperdata: Real-Time Big Data Observability

Pepperdata provides full-stack observability and automated cost optimization for data-intensive workloads on Kubernetes and Amazon EMR, helping Fortune 100 companies reduce infrastructure spend by up to 47%.

‍

A Single Pane of Glass for Massive Clusters

Managing thousands of nodes across hybrid environments requires high-level clarity. We designed the central dashboard to provide an instant health check of the entire big data stack, correlating infrastructure metrics with application performance to surface bottlenecks before they impact SLAs.
‍

Key Features Shown: * Real-time CPU & Memory utilization

Cluster-wide health scoring
Active vs. Pending workload tracking

Deep-Dive Application Diagnostics

When a Spark job fails or runs slow, engineers need to know why. This interface allows users to drill down into individual jobs to see task-level details, data skew, and resource usage. We focused on making these complex data points actionable to drastically reduce Mean Time to Repair (MTTR).
‍

Key Troubleshooting Features:

"Wait-Time" analysis for identifies bottlenecks
Error log correlation with hardware telemetry
Side-by-side job comparisons

Autonomous Cost & Waste Management

Most cloud clusters are over-provisioned. Our design for the Capacity Optimizer highlights "Reclaimable Waste"—showing exactly how much money is being saved by Pepperdata’s autonomous scaling. The UI visualizes the gap between allocated and actual usage in real-time.
‍

Key Features Shown:

Cost savings forecasting (ROI tracking)
Reclaimed resource metrics (vCPU and GB-Hours)
Automated instance-hour reduction charts

Financial Accountability for Big Data

To solve the problem of "runaway" cloud costs, we designed detailed chargeback reports. This allows IT leaders to attribute specific costs to different business units, departments, or individual users, turning a massive cloud bill into a transparent, accountable budget.
‍

Key Features Shown: