A fast on-chip profiler memory

Roman L Lysecky, Susan Cotterell, Frank Vahid

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

Profiling an application executing on a microprocessor is part of the solution to numerous software and hardware optimization and design automation problems. Most current profiling techniques suffer from runtime overhead, inaccuracy, or slowness, and the traditional non-intrusive method of using a logic analyzer doesn't work for today's system-on-a-chip having embedded cores. We introduce a novel on-chip memory architecture that overcomes these limitations. The architecture, which we call ProMem, is based on a pipelined binary tree structure. It achieves single-cycle throughput, so it can keep up with today's fastest pipelined processors. It can also be laid out efficiently and scales very well, becoming more efficient the larger it gets. The memory can be used in a wide-variety of common profiling situations, such as instruction profiling, value profiling, and network traffic profiling, which in turn can be used to guide numerous design automation tasks.

Original languageEnglish (US)
Title of host publicationProceedings - Design Automation Conference
Pages28-33
Number of pages6
StatePublished - 2002
Externally publishedYes
Event39th Annual Design Automation Conference, DAC'02 - New Orleans, LA, United States
Duration: Jun 10 2002Jun 14 2002

Other

Other39th Annual Design Automation Conference, DAC'02
CountryUnited States
CityNew Orleans, LA
Period6/10/026/14/02

Fingerprint

Automation
Data storage equipment
Memory architecture
Binary trees
Microprocessor chips
Throughput
Hardware

Keywords

  • Adaptive architectures
  • Binary tree
  • Embedded CAD
  • Embedded systems
  • Low power
  • Memory design
  • Platform tuning
  • Profiling
  • System-on-a-chip

ASJC Scopus subject areas

  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Lysecky, R. L., Cotterell, S., & Vahid, F. (2002). A fast on-chip profiler memory. In Proceedings - Design Automation Conference (pp. 28-33)

A fast on-chip profiler memory. / Lysecky, Roman L; Cotterell, Susan; Vahid, Frank.

Proceedings - Design Automation Conference. 2002. p. 28-33.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lysecky, RL, Cotterell, S & Vahid, F 2002, A fast on-chip profiler memory. in Proceedings - Design Automation Conference. pp. 28-33, 39th Annual Design Automation Conference, DAC'02, New Orleans, LA, United States, 6/10/02.
Lysecky RL, Cotterell S, Vahid F. A fast on-chip profiler memory. In Proceedings - Design Automation Conference. 2002. p. 28-33
Lysecky, Roman L ; Cotterell, Susan ; Vahid, Frank. / A fast on-chip profiler memory. Proceedings - Design Automation Conference. 2002. pp. 28-33
@inproceedings{4d54f12dc2c14439afd156fee845a30e,
title = "A fast on-chip profiler memory",
abstract = "Profiling an application executing on a microprocessor is part of the solution to numerous software and hardware optimization and design automation problems. Most current profiling techniques suffer from runtime overhead, inaccuracy, or slowness, and the traditional non-intrusive method of using a logic analyzer doesn't work for today's system-on-a-chip having embedded cores. We introduce a novel on-chip memory architecture that overcomes these limitations. The architecture, which we call ProMem, is based on a pipelined binary tree structure. It achieves single-cycle throughput, so it can keep up with today's fastest pipelined processors. It can also be laid out efficiently and scales very well, becoming more efficient the larger it gets. The memory can be used in a wide-variety of common profiling situations, such as instruction profiling, value profiling, and network traffic profiling, which in turn can be used to guide numerous design automation tasks.",
keywords = "Adaptive architectures, Binary tree, Embedded CAD, Embedded systems, Low power, Memory design, Platform tuning, Profiling, System-on-a-chip",
author = "Lysecky, {Roman L} and Susan Cotterell and Frank Vahid",
year = "2002",
language = "English (US)",
isbn = "1581134614",
pages = "28--33",
booktitle = "Proceedings - Design Automation Conference",

}

TY - GEN

T1 - A fast on-chip profiler memory

AU - Lysecky, Roman L

AU - Cotterell, Susan

AU - Vahid, Frank

PY - 2002

Y1 - 2002

N2 - Profiling an application executing on a microprocessor is part of the solution to numerous software and hardware optimization and design automation problems. Most current profiling techniques suffer from runtime overhead, inaccuracy, or slowness, and the traditional non-intrusive method of using a logic analyzer doesn't work for today's system-on-a-chip having embedded cores. We introduce a novel on-chip memory architecture that overcomes these limitations. The architecture, which we call ProMem, is based on a pipelined binary tree structure. It achieves single-cycle throughput, so it can keep up with today's fastest pipelined processors. It can also be laid out efficiently and scales very well, becoming more efficient the larger it gets. The memory can be used in a wide-variety of common profiling situations, such as instruction profiling, value profiling, and network traffic profiling, which in turn can be used to guide numerous design automation tasks.

AB - Profiling an application executing on a microprocessor is part of the solution to numerous software and hardware optimization and design automation problems. Most current profiling techniques suffer from runtime overhead, inaccuracy, or slowness, and the traditional non-intrusive method of using a logic analyzer doesn't work for today's system-on-a-chip having embedded cores. We introduce a novel on-chip memory architecture that overcomes these limitations. The architecture, which we call ProMem, is based on a pipelined binary tree structure. It achieves single-cycle throughput, so it can keep up with today's fastest pipelined processors. It can also be laid out efficiently and scales very well, becoming more efficient the larger it gets. The memory can be used in a wide-variety of common profiling situations, such as instruction profiling, value profiling, and network traffic profiling, which in turn can be used to guide numerous design automation tasks.

KW - Adaptive architectures

KW - Binary tree

KW - Embedded CAD

KW - Embedded systems

KW - Low power

KW - Memory design

KW - Platform tuning

KW - Profiling

KW - System-on-a-chip

UR - http://www.scopus.com/inward/record.url?scp=0036054544&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036054544&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0036054544

SN - 1581134614

SP - 28

EP - 33

BT - Proceedings - Design Automation Conference

ER -