Sept. 23 — Minimal Metrics has announced an early-adopter program for PerfMiner, their next-generation performance and ROI optimization tool. PerfMiner is an advanced site optimization system for HPC and enterprise-class data centers, providing actionable information on how to optimize a site’s productivity and return on investment in hardware, software, and personnel.
PerfMiner uses state-of-the-art, lightweight, and pervasive performance data collection technology, automates its collection, and mines the data for key performance indicators. These indicators were developed through Minimal Metrics’ extensive experience tuning HPC and enterprise application performance. They are presented in an audience-specific, drill-down hierarchy that provides accountability for site productivity down to the performance of individual application threads.
PerfMiner is currently being used at Sandia National Laboratory to analyze multiple programming models and application performance across a range of architectures in planning for the Exascale systems of tomorrow. Scania AB, a major industrial equipment manufacturer, is evaluating it to reduce commercial software licensing costs and hold vendors accountable for promised performance.
“Moore’s law may be dead, but Mucci’s law is going strong,” said Philip Mucci, co-founder and President of Minimal Metrics. “Every year, new hardware, middleware applications, and even personnel come with new challenges, and the number of ways to kill performance doubles.”
In what is a vast departure from other systems, PerfMiner measures applications, platforms, and systems from the inside out. It’s the first product of its kind to present an elegant comprehensive BI-style system for computational resources — one that harvests, processes, and derives the necessary intelligence from both low-level and high-level performance and configuration data.
While a number of disparate tools can be used to monitor system health, application performance, and batch system throughput, they provide little to no information sharing across these highly interrelated domains. PerfMiner’s visual cues facilitate ease of interpretation, faster decision-making, and improved allocation of resources. The tool is completely transparent to the user, has negligible overhead, works on applications written in any language, and does not require any modification to the code or operating systems.
“PerfMiner is not just another dashboard of indicators a human must learn to interpret,” said Dr. Tony Castaldo, Chief Data Scientist at Minimal Metrics. “By using the latest techniques in statistics, data mining, and machine learning we identify performance and configuration issues as well as their solutions. As the gross bottlenecks are identified and alleviated, PerfMiner continues to adapt and refine its criteria to identify additional issues and possible resolutions. Nobody else is doing that.”
The Minimal Metrics team offers an extensive history of performance tuning experience. Co-founder Philip Mucci, who has worked for Thinking Machines, IBM Research, and Lawrence Berkeley National Lab, is the inventor of The Performance Application Programming Interface (PAPI), an interface for accessing detailed processor performance information that is used worldwide.
Years of consulting in HPC across a number of industries led Philip to the realization that a vast majority of data centers and high performance computing installations cannot accurately and quantitatively answer fundamental questions about the performance of their investments. “No one can afford the time to understand an entire system, much less its components or individual application, much less optimize it,” Philip said. “On an institutional basis, these individual decisions cascade into organizational inefficiencies that cost millions of dollars per year in lost revenue and needless expenditures.”
Philip believes that both industry and open source have failed to make performance analysis an integral part of organizations that leverage technical computing. “The prevailing attitude has been that a code or system that runs slow is better than a one that doesn’t run at all,” he said. “After 15 years of software development and about 60 years of collective performance consulting, the Minimal Metrics team realized that there was no product on the market even close to PerfMiner.” Their innovative tool changes the process from being symptomatic to preventative, allowing better decision-making throughout the entire lifecycle of applications and infrastructure, and leading to massive savings and increased throughput.
Minimal Metrics is currently engaging with select companies and partners, as it prepares for the 1.0 release of PerfMiner at Supercomputing 2016 in Salt Lake City, Utah, in November. Interested parties are invited to contact Minimal Metrics and become involved in their early adopter program.
About Minimal Metrics
Minimal Metrics was founded in 2012 by Philip Mucci and Tushar Mohan, each respected veterans of the supercomputing industry with 25 years of hands-on programming experience. The company specializes in the development and optimization of algorithms, architectures, and software in performance-critical environments, ranging from mobile platforms to massively parallel supercomputers. Their previous clients have included industry giants such as Intel, Cray, MIPS, and Texas Instruments, numerous national laboratories, and private industry that deals with massive installations serving IoT analytics pipelines, data acquisition systems, and quantitative finance. For more information: www.minimalmetrics.com/perfminer.
Source: Minimal Metrics