High Performance Realtime Convex Solver for Embedded Systems

نویسندگان

  • Ichitaro Yamazaki
  • Saeid Nooshabadi
  • Jack Dongarra
  • Karush Kuhn
چکیده

Convex optimization solvers for embedded systems find widespread use. This letter presents a novel technique to reduce the run-time of decomposition of KKT matrix for the convex optimization solver for an embedded system, by two orders of magnitude. We use the property that although the KKT matrix changes, some of its block sub-matrices are fixed during the solution iterations and the associated solving instances.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Realtime Particle System Simulation and Rendering in Embedded Systems

The market for games for mobile phones/tablets is probably the fastest growing in the whole computer game industry. Although many of these games feature graphics which were out of reach for these systems not long ago, their quality is still far from what can be reached on modern PCs, and many algorithms used in PCs are a bad fit for these systems. For example, volumetric particle systems are ve...

متن کامل

A preconditioned solver for sharp resolution of multiphase flows at all Mach numbers

A preconditioned five-equation two-phase model coupled with an interface sharpening technique is introduced for simulation of a wide range of multiphase flows with both high and low Mach regimes. Harten-Lax-van Leer-Contact (HLLC) Riemann solver is implemented for solving the discretized equations while tangent of hyperbola for interface capturing (THINC) interface sharpening method is applied ...

متن کامل

Modelling and Simulation of a New Combined Blockoriented Microarchitecture Model for Controllers in Responsive Systems

For embedded systems with high reliability and hard realtime constraints, which are called responsive systems, the main design constraint changes from performance to realtime. The class of blockbased instruction set architectures, primarily designed for superscalar processors, has to be changed for this class of application to integrate part of the operating system. The reason is to increase tr...

متن کامل

Fuzzy convergence structures in the framework of L-convex spaces

In this paper,  fuzzy convergence theory in the framework of $L$-convex spaces is introduced. Firstly, the concept of $L$-convex remote-neighborhood spaces is introduced and it is shown that the  resulting category is isomorphic to that of $L$-convex spaces. Secondly, by means of $L$-convex ideals, the notion of $L$-convergence spaces is introduced and it is proved that the  category of $L$-con...

متن کامل

Real-time and low power embedded ℓ1-optimization solver design

Basis pursuit denoising (BPDN) is an optimization method used in cutting edge computer vision and compressive sensing research. Although hosting a BPDN solver on an embedded platform is desirable because analysis can be performed in real-time, existing solvers are generally unsuitable for embedded implementation due to either poor run-time performance or high memory usage. To address the aforem...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016