Software trace cache

Alex Ramírez; Josep L. Larriba-Pey; Carlos Navarro; Josep Torrellas; Mateo Valero

doi:10.1145/2591635.2667175

Software trace cache

Alex Ramírez, Josep L. Larriba-Pey, Carlos Navarro, Josep Torrellas, Mateo Valero

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we address the important problem of instruction fetch for future wide issue superscalar processors. Our approach focuses on understanding the interaction between software and hardware techniques targeting an increase in the instruction fetch bandwidth. That is the objective, for instance, of the Hardware Trace Cache (HTC).

We design a profile based code reordering technique which targets a maximization of the sequentially of instructions, while still trying to minimize instruction cache misses. We call our software approach, Software Trace Cache (STC).

We evaluate our software approach, and then compare it with the HTC and the combination of both techniques. Our results show that for large codes with few loops and deterministic execution sequences like databases and some SPEC-INT codes, the STC offers similar, or better, results than a HTC. Moreover, when combining the software and hardware approaches, we obtain encouraging results: the STC and a small HTC offer similar performance to a much larger HTC alone.

Original language	English (US)
Title of host publication	ICS 2014 - Proceedings of the 28th ACM InternationaI Conference on Supercomputing
Editors	Utpal Banerjee
Publisher	Association for Computing Machinery
Pages	261-268
Number of pages	8
ISBN (Electronic)	9781450328401
DOIs	https://doi.org/10.1145/2591635.2667175
State	Published - Jun 10 2014
Externally published	Yes
Event	25th ACM International Conference on Supercomputing, ICS 2014 - Munich, Germany Duration: Jun 10 2014 → Jun 13 2014

Publication series

Name	Proceedings of the International Conference on Supercomputing

Other

Other	25th ACM International Conference on Supercomputing, ICS 2014
Country/Territory	Germany
City	Munich
Period	6/10/14 → 6/13/14

ASJC Scopus subject areas

General Computer Science

Online availability

10.1145/2591635.2667175

Library availability

Discover UIUC Full Text

Cite this

Software trace cache. / Ramírez, Alex; Larriba-Pey, Josep L.; Navarro, Carlos et al.
ICS 2014 - Proceedings of the 28th ACM InternationaI Conference on Supercomputing. ed. / Utpal Banerjee. Association for Computing Machinery, 2014. p. 261-268 (Proceedings of the International Conference on Supercomputing).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ramírez, A, Larriba-Pey, JL, Navarro, C, Torrellas, J & Valero, M 2014, Software trace cache. in U Banerjee (ed.), ICS 2014 - Proceedings of the 28th ACM InternationaI Conference on Supercomputing. Proceedings of the International Conference on Supercomputing, Association for Computing Machinery, pp. 261-268, 25th ACM International Conference on Supercomputing, ICS 2014, Munich, Germany, 6/10/14. https://doi.org/10.1145/2591635.2667175

@inproceedings{5c784813b22a43519db01a1b2f6e61b8,

title = "Software trace cache",

abstract = "In this paper we address the important problem of instruction fetch for future wide issue superscalar processors. Our approach focuses on understanding the interaction between software and hardware techniques targeting an increase in the instruction fetch bandwidth. That is the objective, for instance, of the Hardware Trace Cache (HTC).We design a profile based code reordering technique which targets a maximization of the sequentially of instructions, while still trying to minimize instruction cache misses. We call our software approach, Software Trace Cache (STC).We evaluate our software approach, and then compare it with the HTC and the combination of both techniques. Our results show that for large codes with few loops and deterministic execution sequences like databases and some SPEC-INT codes, the STC offers similar, or better, results than a HTC. Moreover, when combining the software and hardware approaches, we obtain encouraging results: the STC and a small HTC offer similar performance to a much larger HTC alone.",

author = "Alex Ram{\'i}rez and Larriba-Pey, {Josep L.} and Carlos Navarro and Josep Torrellas and Mateo Valero",

year = "2014",

month = jun,

day = "10",

doi = "10.1145/2591635.2667175",

language = "English (US)",

series = "Proceedings of the International Conference on Supercomputing",

publisher = "Association for Computing Machinery",

pages = "261--268",

editor = "Utpal Banerjee",

booktitle = "ICS 2014 - Proceedings of the 28th ACM InternationaI Conference on Supercomputing",

address = "United States",

note = "25th ACM International Conference on Supercomputing, ICS 2014 ; Conference date: 10-06-2014 Through 13-06-2014",

}

TY - GEN

T1 - Software trace cache

AU - Ramírez, Alex

AU - Larriba-Pey, Josep L.

AU - Navarro, Carlos

AU - Torrellas, Josep

AU - Valero, Mateo

PY - 2014/6/10

Y1 - 2014/6/10

N2 - In this paper we address the important problem of instruction fetch for future wide issue superscalar processors. Our approach focuses on understanding the interaction between software and hardware techniques targeting an increase in the instruction fetch bandwidth. That is the objective, for instance, of the Hardware Trace Cache (HTC).We design a profile based code reordering technique which targets a maximization of the sequentially of instructions, while still trying to minimize instruction cache misses. We call our software approach, Software Trace Cache (STC).We evaluate our software approach, and then compare it with the HTC and the combination of both techniques. Our results show that for large codes with few loops and deterministic execution sequences like databases and some SPEC-INT codes, the STC offers similar, or better, results than a HTC. Moreover, when combining the software and hardware approaches, we obtain encouraging results: the STC and a small HTC offer similar performance to a much larger HTC alone.

AB - In this paper we address the important problem of instruction fetch for future wide issue superscalar processors. Our approach focuses on understanding the interaction between software and hardware techniques targeting an increase in the instruction fetch bandwidth. That is the objective, for instance, of the Hardware Trace Cache (HTC).We design a profile based code reordering technique which targets a maximization of the sequentially of instructions, while still trying to minimize instruction cache misses. We call our software approach, Software Trace Cache (STC).We evaluate our software approach, and then compare it with the HTC and the combination of both techniques. Our results show that for large codes with few loops and deterministic execution sequences like databases and some SPEC-INT codes, the STC offers similar, or better, results than a HTC. Moreover, when combining the software and hardware approaches, we obtain encouraging results: the STC and a small HTC offer similar performance to a much larger HTC alone.

UR - http://www.scopus.com/inward/record.url?scp=84907982923&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907982923&partnerID=8YFLogxK

U2 - 10.1145/2591635.2667175

DO - 10.1145/2591635.2667175

M3 - Conference contribution

AN - SCOPUS:84907982923

T3 - Proceedings of the International Conference on Supercomputing

SP - 261

EP - 268

BT - ICS 2014 - Proceedings of the 28th ACM InternationaI Conference on Supercomputing

A2 - Banerjee, Utpal

PB - Association for Computing Machinery

T2 - 25th ACM International Conference on Supercomputing, ICS 2014

Y2 - 10 June 2014 through 13 June 2014

ER -

Software trace cache

Abstract

Publication series

Other

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this