Pepperdata Introduces Observability and Optimization for GPUs Running Big Data Applications
August 17, 2021
Share this

Pepperdata announced that the Pepperdata product portfolio now includes the ability to monitor Graphics Processing Units (GPUs) running big data applications like Spark on Kubernetes.

Workloads that harness tremendous amounts of data, such as machine learning (ML) and artificial intelligence (AI) applications, require GPUs, which were originally designed to accelerate graphics rendering. That extra processing power comes with a high price tag, and it requires near-constant monitoring for resource waste to get the best performance at the lowest possible cost.

Pepperdata now monitors GPU performance, providing the visibility needed for Spark applications running on Kubernetes and utilizing the processing power of GPUs. With this new visibility, companies can improve the performance of their Spark apps running on those GPUs and manage costs at a granular level.

Unlike traditional infrastructure monitoring, which is limited to the platform, the Pepperdata solution provides visibility into GPU resource utilization at the application level. Pepperdata also provides instant recommendations for optimization. Features include:

- Visibility into GPU memory usage and waste

- Fine-tuning of GPU usage through end-user recommendations

- Ability to attribute usage and cost to specific end-users

“Spark on Kubernetes is quickly becoming a dominant part of the compute infrastructure as data-intensive ML and AI applications proliferate,” said Ash Munshi, CEO, Pepperdata. “GPUs can handle these workloads, but they are expensive to buy and are power-intensive. Until now, there hasn’t been a way to view and manage the infrastructure and applications, which can lead to unnecessary waste and overspending for big data workloads. With Pepperdata, organizations can properly size their GPU hardware investments and have the confidence that they are utilizing them well.”

There are products on the market for monitoring GPUs, but they typically lack long-term storage, the ability to scale, and often do not correlate infrastructure metrics to applications. Pepperdata solves these problems with insight for data center operators, data scientists, and ML/AI developers. They can now understand who is using what resources, optimize to eliminate waste so jobs can be tuned and prioritized, and make sure costs are assigned appropriately to the right users or groups across the enterprise.

Share this

The Latest

June 29, 2022

When it comes to AIOps predictions, there's no question of AI's value in predictive intelligence and faster problem resolution for IT teams. In fact, Gartner has reported that there is no future for IT Operations without AIOps. So, where is AIOps headed in five years? Here's what the vendors and thought leaders in the AIOps space had to share ...

June 27, 2022

A new study by OpsRamp on the state of the Managed Service Providers (MSP) market concludes that MSPs face a market of bountiful opportunities but must prepare for this growth by embracing complex technologies like hybrid cloud management, root cause analysis and automation ...

June 27, 2022

Hybrid work adoption and the accelerated pace of digital transformation are driving an increasing need for automation and site reliability engineering (SRE) practices, according to new research. In a new survey almost half of respondents (48.2%) said automation is a way to decrease Mean Time to Resolution/Repair (MTTR) and improve service management ...

June 23, 2022

Digital businesses don't invest in monitoring for monitoring's sake. They do it to make the business run better. Every dollar spent on observability — every hour your team spends using monitoring tools or responding to what they reveal — should tie back directly to business outcomes: conversions, revenues, brand equity. If they don't? You might be missing the forest for the trees ...

June 22, 2022

Every day, companies are missing customer experience (CX) "red flags" because they don't have the tools to observe CX processes or metrics. Even basic errors or defects in automated customer interactions are left undetected for days, weeks or months, leading to widespread customer dissatisfaction. In fact, poor CX and digital technology investments are costing enterprises billions of dollars in lost potential revenue ...

June 21, 2022

Organizations are moving to microservices and cloud native architectures at an increasing pace. The primary incentive for these transformation projects is typically to increase the agility and velocity of software release and product innovation. These dynamic systems, however, are far more complex to manage and monitor, and they generate far higher data volumes ...

June 16, 2022

Global IT teams adapted to remote work in 2021, resolving employee tickets 23% faster than the year before as overall resolution time for IT tickets went down by 7 hours, according to the Freshservice Service Management Benchmark Report from Freshworks ...

June 15, 2022

Once upon a time data lived in the data center. Now data lives everywhere. All this signals the need for a new approach to data management, a next-gen solution ...

June 14, 2022

Findings from the 2022 State of Edge Messaging Report from Ably and Coleman Parkes Research show that most organizations (65%) that have built edge messaging capabilities in house have experienced an outage or significant downtime in the last 12-18 months. Most of the current in-house real-time messaging services aren't cutting it ...

June 13, 2022
Today's users want a complete digital experience when dealing with a software product or system. They are not content with the page load speeds or features alone but want the software to perform optimally in an omnichannel environment comprising multiple platforms, browsers, devices, and networks. This calls into question the role of load testing services to check whether the given software under testing can perform optimally when subjected to peak load ...