Profiling Halide DSL with CPU Performance Events for Schedule Optimization

2019 ◽  
Author(s):  
Rafael Ravedutti L. Machado ◽  
André Murbach Maidl ◽  
Daniel Weingaertner
Keyword(s):  
Schedule Optimization ◽  
Performance Events

scholarly journals Microarchitectural Characterization on a Mobile Workload

2021 ◽  
Vol 11 (3) ◽  
pp. 1225
Author(s):  
Woohyong Lee ◽  
Jiyoung Lee ◽  
Bo Kyung Park ◽  
R. Young Chul Kim
Keyword(s):  
Performance Monitoring ◽  
Performance Metrics ◽  
Performance Comparison ◽  
Instruction Level Parallelism ◽  
Data Set ◽  
Performance Events ◽  
Hardware Performance Counters ◽  
On Chip ◽  
The Comparative Study ◽  
Level Parallelism

Geekbench is one of the most referenced cross-platform benchmarks in the mobile world. Most of its workloads are synthetic but some of them aim to simulate real-world behavior. In the mobile world, its microarchitectural behavior has been reported rarely since the hardware profiling features are limited to the public. As a popular mobile performance workload, it is hard to find Geekbench’s microarchitecture characteristics in mobile devices. In this paper, a thorough experimental study of Geekbench performance characterization is reported with detailed performance metrics. This study also identifies mobile system on chip (SoC) microarchitecture impacts, such as the cache subsystem, instruction-level parallelism, and branch performance. After the study, we could understand the bottleneck of workloads, especially in the cache sub-system. This means that the change of data set size directly impacts performance score significantly in some systems and will ruin the fairness of the CPU benchmark. In the experiment, Samsung’s Exynos9820-based platform was used as the tested device with Android Native Development Kit (NDK) built binaries. The Exynos9820 is a superscalar processor capable of dual issuing some instructions. To help performance analysis, we enable the capability to collect performance events with performance monitoring unit (PMU) registers. The PMU is a set of hardware performance counters which are built into microprocessors to store the counts of hardware-related activities. Throughout the experiment, functional and microarchitectural performance profiles were fully studied. This paper describes the details of the mobile performance studies above. In our experiment, the ARM DS5 tool was used for collecting runtime PMU profiles including OS-level performance data. After the comparative study is completed, users will understand more about the mobile architecture behavior, and this will help to evaluate which benchmark is preferable for fair performance comparison.

The Optimization of Project Time-Cost Based on Particle Swarm Optimization Algorithm

2013 ◽  
Vol 791-793 ◽  
pp. 1423-1426
Author(s):  
Hai Min Wei ◽  
Rong Guang Liu
Keyword(s):  
Particle Swarm Optimization ◽  
Particle Swarm Optimization Algorithm ◽  
Cost Optimization ◽  
Particle Swarm ◽  
Implementation Process ◽  
Pso Algorithm ◽  
Project Schedule ◽  
Swarm Optimization ◽  
Schedule Optimization ◽  
Schedule Management

Project schedule management is the management to each stage of the degree of progress and project final deadline in the project implementation process. Its purpose is to ensure that the project can meet the time constraints under the premise of achieving its overall objectives.When the progress of schedule found deviation in the process of schedule management ,the progress of the plan which have be advanced previously need to adjust.This article mainly discussed to solve the following two questions:establish the schedule optimization model by using the method of linear;discuss the particle swarm optimization (PSO) algorithm and its parameters which have effect on the algorithm:Particle swarm optimization (PSO) algorithm is presented in the time limited project and the application of a cost optimization.

Optimization of the Receipt and Delivery Schedules for a Transfer Tank Farm of Products Pipeline Network

2017 ◽  
Author(s):  
Li Wang ◽  
Changchun Wu ◽  
Lili Zuo ◽  
Yanfei Huang ◽  
Haihong Chen
Keyword(s):  
Material Balance ◽  
Optimal Solution ◽  
Planning Horizon ◽  
Oil Products ◽  
Mixed Integer ◽  
Pipeline Network ◽  
Schedule Optimization ◽  
Tank Farm ◽  
Dynamic Time ◽  
Products Pipeline

Transfer tank farms play an important role in an oil products pipeline network, which receive oil products from upstream pipelines and deliver them to downstream pipelines. The scheduling problem for oil products supply chain is very complicated because of numerous constraints to be considered. The published literatures on schedule optimization of oil products pipeline network usually focus on the batch plans of each pipeline, without consideration on the receipt and delivery schedule of transfer tank farm. In this paper, a mixed-integer linear programming (MILP) model is developed for the schedule optimization of transfer tank farm. The objective of the model is to minimize switching times of the tank operations of a tank farm during a planning horizon, while fulfilling the products transmission requirements of the upstream and downstream pipelines of the tank farm. The constraints of the model include material balance, the operational rules of tanks, the topological structure constraints of the tank farm, the settling period of the oil products stored in dedicated tank and so on. To satisfy the constraint of fulfilling the specific transmission requirements of pipelines, concepts of static and dynamic time slot are proposed. A continuous time representation is used to obtain accurate optimal schedules and decrease scale of the model by reducing the number of variables. The model is solved by CPLEX solver for a transfer tank farm of an oil products pipeline network in China. Some examples are tested under different scenarios and the results show that global optimal solution can be obtain at acceptable computational costs.

CSP-Driven Schedule Optimization in NC Equipments Grid Considering Complex Correlated Process Flows

2010 ◽  
Vol 97-101 ◽  
pp. 2650-2653
Author(s):  
Xin Lin ◽  
Ya Bo Luo
Keyword(s):  
Boundary Conditions ◽  
Numerical Control ◽  
Just In Time ◽  
Grid Technology ◽  
Np Hard ◽  
Potential Approach ◽  
Schedule Optimization ◽  
Grid Workflow ◽  
Optimum Solution

Numerical Control (NC) equipments sharing over Internet is a potential approach for improving the utilization of capacity, which is still a complex NP hard difficulty due to the complexity of schedule. This research takes the Constraints Satisfied Problem (CSP) thinking to solve the above problem employing grid technology. First, regarding the NC equipments sharing as CSP, the CSP model for NC equipments grid is developed taking the flexible constraints as the optimum objective and the rigid constraints as the boundary conditions. Second, the detailed algorithm and steps for optimization of NC equipments grid workflow based on Just-In-Time (JIT) thinking are proposed. Finally the digital experiment demonstrates the advantages of the CSP-driven methodology that has the higher efficiency for searching optimum solution.

Sign in / Sign up

Export Citation Format

Share Document