Performance Tuning Research Articles

Implementation of a Direct Fuzzy Controller Applied to a Helicopter with one Degree of Freedom

Conventional PID controllers are used in linear processes, showing an appropriate performance and simple tuning methods, however, in non-linear processes with changes in the set-point their response is degraded. In recent decades, the use of fuzzy controllers has increased, due its non-linear behavior and an accurate model of the process is not required. In this document, a new model-based tuning methodology is implemented, which has only been tested in a simulation environment and consists in the calculation of the scale factors of the controller from the parameters of the dynamic response of the system. The tuning method was applied to an one degree of freedom helicopter through a direct fuzzy controller. The platform implemented was tested against changes in the set-point and rejection of disturbances. The performance of controllers was measured through the ITAE and IAE performance indexes. The results of implementation and tests shown a better performance of the fuzzy controller regarding the conventional PID controller without complex computational intelligence algorithms or extensive trial and error methods for tuning.

Design and Control of High-Density High-Voltage Smart Converter for Food Ohmic Heating

The design of an innovative converter for ohmic heating of fluid foods is presented. Application-specific requirements are addressed with new solutions and verified on a commercial-size 60-kW prototype. A multioutput high-voltage (0.8 to 4 kV) transformer with automatic tap change enriches a basic H-bridge topology, enabling processing of a wide set of foods and allowing grounding one of the outputs. High power density is achieved by power switches Pareto optimization in the reliability-efficiency objectives and by pushing the switching frequency up to 30 kHz, solving the electrode corrosion problem, too. Reproducible and reliable operation is assured by three different control strategies: load variability is dealt with high-dynamics power control, repeatable output is guaranteed by input voltage compensation, and optimal working point is ensured by control of the transformer primary current dc component. This last control also allows removing the output capacitors, improving density metrics with respect to the state-of-the-art. Analytical models for all control strategies are given to support dynamic performance tuning while guaranteeing stability.

ApproxHPVM: a portable compiler IR for accuracy-aware optimizations

We propose ApproxHPVM, a compiler IR and system designed to enable accuracy-aware performance and energy tuning on heterogeneous systems with multiple compute units and approximation methods. ApproxHPVM automatically translates end-to-end application-level quality metrics into accuracy requirements for individual operations. ApproxHPVM uses a hardware-agnostic accuracy-tuning phase to do this translation that provides greater portability across heterogeneous hardware platforms and enables future capabilities like accuracy-aware dynamic scheduling and design space exploration. ApproxHPVM incorporates three main components: (a) a compiler IR with hardware-agnostic approximation metrics, (b) a hardware-agnostic accuracy-tuning phase to identify error-tolerant computations, and (c) an accuracy-aware hardware scheduler that maps error-tolerant computations to approximate hardware components. As ApproxHPVM does not incorporate any hardware-specific knowledge as part of the IR, it can serve as a portable virtual ISA that can be shipped to all kinds of hardware platforms. We evaluate our framework on nine benchmarks from the deep learning domain and five image processing benchmarks. Our results show that our framework can offload chunks of approximable computations to special-purpose accelerators that provide significant gains in performance and energy, while staying within user-specified application-level quality metrics with high probability. Across the 14 benchmarks, we observe from 1-9x performance speedups and 1.1-11.3x energy reduction for very small reductions in accuracy.

ZIF-8 Membrane Separation Performance Tuning by Vapor Phase Ligand Treatment.

Vapor phase ligand treatment (VPLT) of 2-aminobenzimidazole (2abIm) for 2-methylimidazole (2mIm) in ZIF-8 membranes prepared by two different methods (LIPS: ligand induced permselectivation and RTD: rapid thermal deposition) results in a notable shift of the molecular level cut-off to smaller molecules establishing selectivity improvements from ca. 1.8 to 5 for O2 /N2 ; 2.2 to 32 for CO2 /CH4 ; 2.4 to 24 for CO2 /N2 ; 4.8 to 140 for H2 /CH4 and 5.2 to 126 for H2 /N2 . Stable (based on a one-week test) oxygen-selective air separation performance at ambient temperature, 7 bar(a) feed, and 1 bar(a) sweep-free permeate with a mixture separation factor of 4.5 and oxygen flux of 2.6×10-3 mol m-2 s-1 is established. LIPS and RTD membranes exhibit fast and gradual evolution upon a 2abIm-VPLT, respectively, reflecting differences in their thickness and microstructure. Functional reversibility is demonstrated by showing that the original permeation properties of the VPLT-LIPS membranes can be recovered upon 2mIm-VPLT.

ZIF‐8 Membrane Separation Performance Tuning by Vapor Phase Ligand Treatment

AbstractVapor phase ligand treatment (VPLT) of 2‐aminobenzimidazole (2abIm) for 2‐methylimidazole (2mIm) in ZIF‐8 membranes prepared by two different methods (LIPS: ligand induced permselectivation and RTD: rapid thermal deposition) results in a notable shift of the molecular level cut‐off to smaller molecules establishing selectivity improvements from ca. 1.8 to 5 for O2/N2; 2.2 to 32 for CO2/CH4; 2.4 to 24 for CO2/N2; 4.8 to 140 for H2/CH4 and 5.2 to 126 for H2/N2. Stable (based on a one‐week test) oxygen‐selective air separation performance at ambient temperature, 7 bar(a) feed, and 1 bar(a) sweep‐free permeate with a mixture separation factor of 4.5 and oxygen flux of 2.6×10−3 mol m−2 s−1 is established. LIPS and RTD membranes exhibit fast and gradual evolution upon a 2abIm‐VPLT, respectively, reflecting differences in their thickness and microstructure. Functional reversibility is demonstrated by showing that the original permeation properties of the VPLT‐LIPS membranes can be recovered upon 2mIm‐VPLT.

Managing code transformations for better performance portability

Code optimization is an intricate task that is getting more complex as computing systems evolve. Managing the program optimization process, including the implementation and evaluation of code variants, is tedious, inefficient, and errors are likely to be introduced in the process. Moreover, because each platform typically requires a different sequence of transformations to fully harness its computing power, the optimization process complexity grows as new platforms are adopted. To address these issues, systems and frameworks have been proposed to automate the code optimization process. They, however, have not been widely adopted and are primarily used by experts with deep knowledge about underlying architecture and compiler intricacies. This article describes the requirements that we believe necessary for making automatic performance tuning more broadly used, especially in complex, long-lived high-performance computing applications. Besides discussing limitations of current systems and strategies to overcome these, we describe the design of a system that is able to semi-automatically generate efficient platform-specific code. In the proposed system, the code optimization is programmer-guided, separately from application code, on an external file in what we call optimization programming. The language to program the optimization process is able to represent complex collections of transformations and, as a result, generate efficient platform-specific code. A database manages different optimized versions of code regions, providing a pragmatic approach to performance portability, and the framework itself has separate components, allowing the optimized code to be used on systems without installing all of the modules required for the code generation. We present experiments on two different platforms to illustrate the generation of efficient platform-specific code that performs comparable to hand-optimized, vendor-provided code.

Intelligent Liquid Integrated Functional Entity: A Basic Way to Innovate Future Advanced Biomimetic Soft Robotics

The liquid is an essential component of natural creatures with profound capacities in assisting homeostatic regulation of physiological functions. However, its participation in more diverse biological behaviors, such as intelligence, motion, energy, etc., is seriously neglected and is undisclosed thus far. Herein, it is considered whether liquid systems can offer pivotal hints in the development of intelligent machines. Starting from this core point, a basic conjecture is proposed that specifically equips a robotic system with liquid that brings about revolutionary design of intelligent systems and plays an elementary role in molding future advanced soft robots. Through a comparative analogy between artificially made machines and natural animals, the basic category of the so‐termed Intelligent Liquid Integrated Functional Entity (I‐LIFE) is drafted and thus enabled. Then, functions of such unconventional robotic liquid are expounded from five aspects, including motion, energy supply, material performance tuning, sensing, and intelligence, respectively. Typical application examples are also illustrated. Furthermore, a group of intelligent liquids are recommended as candidates for developing future robotic systems. The fundamental scientific and technological challenges lying behind these recommendations are outlined. Finally, prospects of I‐LIFE are pointed out, which stimulates the increasing design of a future generation of innovative intelligent robotic systems.

Production Application Performance Data Streaming for System Monitoring

In this article, we present an approach to streaming collection of application performance data. Practical application performance tuning and troubleshooting in production high-performance computing (HPC) environments requires an understanding of how applications interact with the platform, including (but not limited to) parallel programming libraries such as Message Passing Interface (MPI). Several profiling and tracing tools exist that collect heavy runtime data traces either in memory (released only at application exit) or on a file system (imposing an I/O load that may interfere with the performance being measured). Although these approaches are beneficial in development stages and post-run analysis, a systemwide and low-overhead method is required to monitor deployed applications continuously. This method must be able to collect information at both the application and system levels to yield a complete performance picture.In our approach, an application profiler collects application event counters. A sampler uses an efficient inter-process communication method to periodically extract the application counters and stream them into an infrastructure for performance data collection. We implement a tool-set based on our approach and integrate it with the Lightweight Distributed Metric Service (LDMS) system, a monitoring system used on large-scale computational platforms. LDMS provides the infrastructure to create and gather streams of performance data in a low overhead manner. We demonstrate our approach using applications implemented with MPI, as it is one of the most common standards for the development of large-scale scientific applications.We utilize our tool-set to study the impact of our approach on an open source HPC application, Nalu. Our tool-set enables us to efficiently identify patterns in the behavior of the application without source-level knowledge. We leverage LDMS to collect system-level performance data and explore the correlation between the system and application events. Also, we demonstrate how our tool-set can help detect anomalies with a low latency. We run tests on two different architectures: a system enabled with Intel Xeon Phi and another system equipped with Intel Xeon processor. Our overhead study shows our method imposes at most 0.5% CPU usage overhead on the application in realistic deployment scenarios.

Synthesis and Performance Tuning of Sm0.2Ce0.8O2−δ Electrolyte for Low Temperature Solid Oxide Fuel Cell Application

The charge transportation in the solid oxide fuel cell electrolyte, Sm0.2Ce0.8O2−δ (SDC); has been elucidated by using DC and AC measurements as a function of grain size at temperature 500°C. Initially, chemically homogeneous pellets of SDC were prepared using its powder synthesized by oxalate co-precipitation method and then mean crystallite-size of the SDC samples was varied by adjusting the sintering temperature. The mean crystallite-size was calculated from x-ray diffraction data by using the Debye–Scherrer equation. Further, the samples were examined for their crystal structure, crystallite-size and chemical homogeneity. Electrochemical impedance spectroscopy was used to understand electrical properties of the samples and its correlation with crystallite-size was revealed. SDC samples having larger crystallites exhibited higher electrical conductivity by providing a number of mobile oxygen ions for conduction. However, a lesser number of oxygen ion vacancies across crystallite-boundaries become a hurdle for oxygen migration through samples having small crystallite-size.

Performance tuning for machine learning-based software development effort prediction models

Software development effort estimation is a critical activity of the project management process. In this study, machine learning algorithms were investigated in conjunction with feature transformation, feature selection, and parameter tuning techniques to estimate the development effort accurately and a new model was proposed as part of an expert system. We preferred the most general-purpose algorithms, applied parameter optimization technique (Grid- Search), feature transformation techniques (binning and one-hot-encoding), and feature selection algorithm (principal component analysis). All the models were trained on the ISBSG datasets and implemented by using the scikit-learn package in the Python language. The proposed model uses a multilayer perceptron as its underlying algorithm, applies binning of the features to transform continuous features and one-hot-encoding technique to transform categorical data into numerical values as feature transformation techniques, does feature selection based on the principal component analysis method, and performs parameter tuning based on the GridSearch algorithm. We demonstrate that our effort prediction model mostly outperforms the other existing models in terms of prediction accuracy based on the mean absolute residual parameter.

A Framework for Automated Cellular Network Tuning With Reinforcement Learning

Tuning cellular network performance against always occurring wireless impairments can dramatically improve reliability to end users. In this paper, we formulate cellular network performance tuning as a reinforcement learning (RL) problem and provide a solution to improve the performance for indoor and outdoor environments. By leveraging the ability of Q-learning to estimate future performance improvement rewards, we propose two algorithms: (1) closed loop power control (PC) for downlink voice over LTE (VoLTE) and (2) self-organizing network (SON) fault management. The VoLTE PC algorithm uses RL to adjust the indoor base station transmit power so that the signal to interference plus noise ratio (SINR) of a user equipment (UE) meets the target SINR. It does so without the UE having to send power control requests. The SON fault management algorithm uses RL to improve the performance of an outdoor base station cluster by resolving faults in the network through configuration management. Both algorithms exploit measurements from the connected users, wireless impairments, and relevant configuration parameters to solve a non-convex performance optimization problem using RL. Simulation results show that our proposed RL based algorithms outperform the industry standards today in realistic cellular communication environments.

A Variated Monte Carlo Tree Search Algorithm for Automatic Performance Tuning to Achieve Load Scalability in InnoDB Storage Engines

A Variated Monte Carlo Tree Search Algorithm for Automatic Performance Tuning to Achieve Load Scalability in InnoDB Storage Engines

Performance tuning of microfluidic flow-focusing droplet generators.

The required step in all droplet-based devices is droplet formation. A droplet generator must deliver an application-specific performance that includes a prescribed droplet size and generation frequency while producing monodisperse droplets. The desired performance is usually reached through several cost- and time-inefficient design iterations. To address this, we take advantage of a low-cost rapid prototyping method and provide a framework that enables researchers to make informed decisions on how to change geometric parameters and flow conditions to tune the performance of a microfluidic flow-focusing droplet generator. We present the primary and secondary parameters necessary for fine-tuning droplet formation over a wide range of capillary numbers and flow rate ratios. Once the key parameters are identified, we demonstrate the effect of geometric parameters and flow conditions on droplet size, generation rate, polydispersity, and generation regime. Using this framework, a wide range of droplet diameters (i.e., 30-400 μm) and generation rates (i.e., 0.5-800 Hz) was achieved.

Towards AI-Powered Multiple Cloud Management

Cloud users worldwide are looking at the next generation of artificial intelligence (AI) powered cloud management tools to automate cloud performance tuning and anomaly detection. To be effective across clouds, AI tools need a common representation of cloud services and support for machine learning optimization targeting multiple objectives. We put forward the notion that ontology-based models can support both.

An Autotuning Framework for Scalable Execution of Tiled Code via Iterative Polyhedral Compilation

On modern many-core CPUs, performance tuning against complex memory subsystems and scalability for parallelism is mandatory to achieve their potential. In this article, we focus on loop tiling, which plays an important role in performance tuning, and develop a novel framework that analytically models the load balance and empirically autotunes unpredictable cache behaviors through iterative polyhedral compilation using LLVM/Polly. From an evaluation on many-core CPUs, we demonstrate that our autotuner achieves a performance superior to those that use conventional static approaches and well-known autotuning heuristics. Moreover, our autotuner achieves almost the same performance as a brute-force search-based approach.

The MPI_T events interface: An early evaluation and overview of the interface

Understanding the behavior of parallel applications that use the Message Passing Interface (MPI) is critical for optimizing communication performance. Performance tools for MPI currently rely on the PMPI Profiling Interface or the MPI Tool Information Interface, MPI_T, for portably collecting information for performance measurement and analysis. While tools using these interfaces have proven to be extremely valuable for performance tuning, these interfaces only provide synchronous information, i.e., when an MPI or an MPI_T function is called. There is currently no option for collecting information about asynchronous events from within the MPI library. In this work we propose a callback-driven interface for event notification from MPI implementations. Our approach is integrated in the existing MPI_T interface and provides a portable API for tools to discover and register for events of interest. We implement our MPI_T Events interface in Open MPI and demonstrate its functionality and usability with a small logging tool (MEL) as well as an early integration into the comprehensive measurement infrastructure Score-P.

Reproducibility, accuracy and performance of the Feltor code and library on parallel computer architectures

Feltor is a modular and free scientific software package. It allows developing platform independent code that runs on a variety of parallel computer architectures ranging from laptop CPUs to multi-GPU distributed memory systems. Feltor consists of both a numerical library and a collection of application codes built on top of the library. Its main targets are two- and three-dimensional drift- and gyro-fluid simulations with discontinuous Galerkin methods as the main numerical discretization technique.We observe that numerical simulations of a recently developed gyro-fluid model produce non-deterministic results in parallel computations. First, we show how we restore accuracy and bitwise reproducibility algorithmically and programmatically. In particular, we adopt an implementation of the exactly rounded dot product based on long accumulators, which avoids accuracy losses especially in parallel applications. However, reproducibility and accuracy alone fail to indicate correct simulation behavior. In fact, in the physical model slightly different initial conditions lead to vastly different end states. This behavior translates to its numerical representation. Pointwise convergence, even in principle, becomes impossible for long simulation times. We briefly discuss alternative methods to ensure the correctness of results like the convergence of reduced physical quantities of interest, ensemble simulations, invariants or reduced simulation times.In a second part, we explore important performance tuning considerations. We identify latency and memory bandwidth as the main performance indicators of our routines. Based on these, we propose a parallel performance model that predicts the execution time of algorithms implemented in Feltor and test our model on a selection of parallel hardware architectures. We are able to predict the execution time with a relative error of less than 25% for problem sizes between 10−1 and 103 MB. Finally, we find that the product of latency and bandwidth gives a minimum array size per compute node to achieve a scaling efficiency above 50% (both strong and weak).

Performance Tuning of Very Large Scale Integration Interconnects Integrated with Deep Sub Micron Repeaters

Performance Tuning of Very Large Scale Integration Interconnects Integrated with Deep Sub Micron Repeaters

Similarity Metrics for SQL Query Clustering

Database access logs are the starting point for many forms of database administration, from database performance tuning, to security auditing, to benchmark design, and many more. Unfortunately, query logs are also large and unwieldy, and it can be difficult for an analyst to extract broad patterns from the set of queries found therein. Clustering is a natural first step towards understanding the massive query logs. However, many clustering methods rely on the notion of pairwise similarity, which is challenging to compute for SQL queries, especially when the underlying data and database schema is unavailable. We investigate the problem of computing similarity between queries, relying only on the query structure. We conduct a rigorous evaluation of three query similarity heuristics proposed in the literature applied to query clustering on multiple query log datasets, representing different types of query workloads. To improve the accuracy of the three heuristics, we propose a generic feature engineering strategy, using classical query rewrites to standardize query structure. The proposed strategy results in a significant improvement in the performance of all three similarity heuristics.

Physical-Layer Adaptive Resource Allocation in Software-Defined Data Center Networks

With the deep physical-layer programmability enabled by software-defined networking (SDN), it is now feasible to realize optical data center transceivers that adapt their transmission parameters in response to timevarying traffic demand and signal quality conditions. In this paper, we develop a cross-layer performance tuning framework for wavelength-tunable data center transceivers, combining scalable and secure modulation order and code rate adaptation. We develop new physical-layer control modules and combine SDN control with distributed preamble encoding in order to achieve both rate adaptation and node synchronization in a wavelength-routing data center testbed. Our experimental and theoretical studies based on pulse amplitude modulation and low-density parity-check coding point to the significance of joint modulation order and code-rate adaptation in data centers. We report real-time resource adaptation with switching speeds on the order of hundreds of milliseconds.