scholarly journals Investigating Memory System Energy Behavior Using Software and Hardware Optimizations

VLSI Design ◽  
2001 ◽  
Vol 12 (2) ◽  
pp. 151-165 ◽  
Author(s):  
G. Esakkimuthu ◽  
H. S. Kim ◽  
M. Kandemir ◽  
N. Vijaykrishnan ◽  
M. J. Irwin
Keyword(s):  
Compiler Optimization ◽  
Memory System ◽  
Quantitative Comparison ◽  
Optimization Techniques ◽  
Compiler Optimizations ◽  
Cache Optimization ◽  
Energy Behavior ◽  
Software And Hardware ◽  
System Energy ◽  
Hardware Optimizations

Memory system usually consumes a significant amount of energy in many battery-operated devices. In this paper, we provide a quantitative comparison and evaluation of the interaction of two hardware cache optimization mechanisms and three widely used compiler optimization techniques used to reduce the memory system energy. Our presentation is in two parts. First, we focus on a set of memory-intensive benchmark codes and investigate their memory system energy behavior due to data accesses under hardware and compiler optimizations. Then, using four motion estimation codes, we look at the influence of compiler optimizations on the memory system energy considering the overall impact of instruction and data accesses.

scholarly journals Compiler Optimization Techniques for OpenMP Programs

2001 ◽  
Vol 9 (2-3) ◽  
pp. 131-142 ◽  
Author(s):  
Shigehisa Satoh ◽  
Kazuhiro Kusano ◽  
Mitsuhisa Sato
Keyword(s):  
Compiler Optimization ◽  
Distributed Shared Memory ◽  
Memory Systems ◽  
Optimization Techniques ◽  
Dataflow Analysis ◽  
Compiler Optimizations ◽  
Data Dependence Analysis ◽  
Parallel Loops ◽  
Intensive Programs ◽  
Software Distributed Shared Memory

We have developed compiler optimization techniques for explicit parallel programs using the OpenMP API. To enable optimization across threads, we designed dataflow analysis techniques in which interactions between threads are effectively modeled. Structured description of parallelism and relaxed memory consistency in OpenMP make the analyses effective and efficient. We developed algorithms for reaching definitions analysis, memory synchronization analysis, and cross-loop data dependence analysis for parallel loops. Our primary target is compiler-directed software distributed shared memory systems in which aggressive compiler optimizations for software-implemented coherence schemes are crucial to obtaining good performance. We also developed optimizations applicable to general OpenMP implementations, namely redundant barrier removal and privatization of dynamically allocated objects. Experimental results for the coherency optimization show that aggressive compiler optimizations are quite effective for a shared-write intensive program because the coherence-induced communication volume in such a program is much larger than that in shared-read intensive programs.

Tuning garbage collection for reducing memory system energy in an embedded java environment

2002 ◽  
Vol 1 (1) ◽  
pp. 27-55 ◽  
Author(s):  
G. Chen ◽  
R. Shetty ◽  
M. Kandemir ◽  
N. Vijaykrishnan ◽  
M. J. Irwin ◽  
...  
Keyword(s):  
Garbage Collection ◽  
Memory System ◽  
System Energy

scholarly journals Lost In Translation: Exposing Hidden Compiler Optimization Opportunities

2020 ◽  
Author(s):  
Kyriakos Georgiou ◽  
Zbigniew Chamski ◽  
Andres Amaya Garcia ◽  
David May ◽  
Kerstin Eder
Keyword(s):  
Compiler Optimization ◽  
Optimization Techniques ◽  
Raspberry Pi ◽  
Routine Testing ◽  
New Approach ◽  
Development Cycle ◽  
Iterative Compilation ◽  
Dependent Potential ◽  
Lost In Translation

Abstract Existing iterative compilation and machine learning-based optimization techniques have been proven very successful in achieving better optimizations than the standard optimization levels of a compiler. However, they were not engineered to support the tuning of a compiler’s optimizer as part of the compiler’s daily development cycle. In this paper, we first establish the required properties that a technique must exhibit to enable such tuning. We then introduce an enhancement to the classic nightly routine testing of compilers, which exhibits all the required properties and thus is capable of driving the improvement and tuning of the compiler’s common optimizer. This is achieved by leveraging resource usage and compilation information collected while systematically exploiting prefixes of the transformations applied at standard optimization levels. Experimental evaluation using the LLVM v6.0.1 compiler demonstrated that the new approach was able to reveal hidden cross-architecture and architecture-dependent potential optimizations on two popular processors: the Intel i5-6300U and the Arm Cortex-A53-based Broadcom BCM2837 used in the Raspberry Pi 3B+. As a case study, we demonstrate how the insights from our approach enabled us to identify and remove a significant shortcoming of the CFG simplification pass of the LLVM v6.0.1 compiler.

Forming a TED talks sphere for convenient search

2020 ◽  
Vol 38 (2) ◽  
pp. 403-420
Author(s):  
Chia-Hui Shih ◽  
Han-Lin Li ◽  
Chih-Chien Hu ◽  
Bertrand M.T. Lin
Keyword(s):  
Phase 1 ◽  
Optimization Method ◽  
Optimization Techniques ◽  
Content Type ◽  
Video Systems ◽  
Software And Hardware ◽  
Survey And Evaluation ◽  
Five Phases ◽  
User Friendly ◽  
Ted Talks

Purpose TED (Technology, Entertainment, Design, www.ted.com/) Talks has been one of the most popular video systems. However, the current TED Talks system expressed its inquired videos as in a two-dimensional (2D) table, which is inconvenient for searching the relationships among videos and tags. This study converts the TED Talks table into a sphere by using optimization techniques to help users search for preferred videos. Design/methodology/approach There are five phases in this study as follows. Phase 1: Reorganize data of 36 tags and 108 videos; Phase 2: Allocate tags on the TED sphere; Phase 3: Allocate videos on the TED sphere; Phase 4: Develop an online interactive TED retrieval system; and Phase 5: Perform survey and evaluation. Findings One survey demonstrated that the TED Talks sphere is more convenient for searching videos, as it is more user-friendly because of its graphical user interface, more convenient to use, more useful for retrieving information and can facilitate a more responsive search for users’ preferred videos. Research limitations/implications The numbers of tags and videos able to be displayed on a sphere is limited by the capacity of an optimization software and hardware. Practical implications The proposed sphere system can be used by a large number of users of TED Talks groups. Social implications This sphere systems can also be applied to other fields which use 2D forms to display the relationships among objects. Originality/value This study uses an optimization method to convert a 2D form into a 3D sphere to highlight the relationships among numerous objects.

Sign in / Sign up

Export Citation Format

Share Document