A Conceptual Framework for HPC Operational Data Analytics

Netti, Alessio; Shin, Woong; Ott, Michael; Wilde, Torsten; Bates, Natalie

doi:10.1109/cluster48925.2021.00086

Cited by 13 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, several works have been proposed to extend the power monitoring attainable from the voltage regulator modules leveraging shunt resistors, current probes, and out-ofband telemetry [23]. In addition, Operational Data Analytics [26] (ODA) has been introduced focusing on monitoring and managing large scale HPC installations. In this area, vertical solutions encompassing all layers (from data gathering and storage to processing and analysis) have been proposed.…”

Section: Related Workmentioning

confidence: 99%

Monte Cimone: Paving the Road for the First Generation of RISC-V High-Performance Computers

Bartolini¹,

Ficarelli²,

Parisi³

et al. 2022

Preprint

View full text Add to dashboard Cite

The new open and royalty-free RISC-V ISA is attracting interest across the whole computing continuum, from microcontrollers to supercomputers. High-performance RISC-V processors and accelerators have been announced, but RISC-Vbased HPC systems will need a holistic co-design effort, spanning memory, storage hierarchy interconnects and full software stack. In this paper, we describe Monte Cimone, a fully-operational multi-blade computer prototype and hardware-software test-bed based on U740, a double-precision capable multi-core, 64-bit RISC-V SoC. Monte Cimone does not aim to achieve strong floating-point performance, but it was built with the purpose of "priming the pipe" and exploring the challenges of integrating a multi-node RISC-V cluster capable of providing an HPC production stack including interconnect, storage and power monitoring infrastructure on RISC-V hardware. We present the results of our hardware/software integration effort, which demonstrate a remarkable level of software and hardware readiness and maturity -showing that the first generation of RISC-V HPC machines may not be so far in the future.

show abstract

Section: Related Workmentioning

confidence: 99%

Monte Cimone: Paving the Road for the First Generation of RISC-V High-Performance Computers

Bartolini¹,

Ficarelli²,

Parisi³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Monitoring the health of all those subsystems is an increasingly daunting task for system administrators. To simplify this monitoring task and reduce the time between anomaly insurgency and response by the administrators, automatic anomaly detection systems have been introduced in recent years [3].…”

Section: Introductionmentioning

confidence: 99%

“…Modern supercomputers are endowed with monitoring systems that give the system administrators a holistic view of the system [3]. Data collected by these monitoring systems and historical data describing system availability are the basis for Ma-chine Learning anomaly detection approaches [6,7,8,9,10], which build data-driven models of the supercomputer and its computing nodes.…”

Section: Introductionmentioning

confidence: 99%

“…Downtime tracking also has to be able to record failures with the same granularity as the other monitoring services. Some methods in production HPC systems only record downtime events by date [1,2,3]. In most production HPC systems, accurate anomaly detection is thus not readily achievable.…”

Section: Introductionmentioning

confidence: 99%

“…It would then learn online and without any interaction with the system administrators. Additionally, such a system would be easier to deploy as it would require no additional framework to report and record anomalous events (in addition to the monitoring infrastructure needed to build the data-driven model of the target supercomputer -a type of infrastructure which is becoming more and more widespread in current HPC facilities [3]).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Ruad: Unsupervised Anomaly Detection in Hpc Systems

Molan¹,

Borghesi

Cesarini³

et al. 2022

SSRN Journal

View full text Add to dashboard Cite

Analysing Supercomputer Nodes Behaviour with the Latent Representation of Deep Learning Models

Molan

Borghesi

Benini

et al. 2022

Euro-Par 2022: Parallel Processing

View full text Add to dashboard Cite

Anomaly detection systems are vital in ensuring the availability of modern High-Performance Computing (HPC) systems, where many components can fail or behave wrongly. Building a data-driven representation of the computing nodes can help with predictive maintenance and facility management. Luckily, most of the current supercomputers are endowed with monitoring frameworks that can build such representations in conjunction with Deep Learning (DL) models. In this work, we propose a novel semi-supervised DL approach based on autoencoder networks and clustering algorithms (applied to the latent representation) to build a digital twin of the computing nodes of the system. The DL model projects the node features into a lower-dimensional space. Then, clustering is applied to capture and reveal underlying, non-trivial correlations between the features. The extracted information provides valuable insights for system administrators and managers, such as anomaly detection and node classification based on their behaviour and operative conditions. We validated the approach on 240 nodes from the Marconi 100 system, a Tier-0 supercomputer located in CINECA (Italy), considering a 10-month period.

show abstract

A Conceptual Framework for HPC Operational Data Analytics

Cited by 13 publications

References 53 publications

Monte Cimone: Paving the Road for the First Generation of RISC-V High-Performance Computers

Monte Cimone: Paving the Road for the First Generation of RISC-V High-Performance Computers

Ruad: Unsupervised Anomaly Detection in Hpc Systems

Analysing Supercomputer Nodes Behaviour with the Latent Representation of Deep Learning Models

Contact Info

Product

Resources

About