AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers

Nayana Prasad Nagendra; Grant Ayers; David I. August; Hyoun Kyu Cho; Svilen Kanev; Christos Kozyrakis; Trivikram Krishnamurthy; Heiner Litz; Tipp Moseley; Parthasarathy Ranganathan

doi:10.1109/MM.2020.2986212

AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers

Nayana Prasad Nagendra
, Grant Ayers
, David I. August
, Hyoun Kyu Cho
, Svilen Kanev
, Christos Kozyrakis
, Trivikram Krishnamurthy
, Heiner Litz
, Tipp Moseley
, Parthasarathy Ranganathan

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

It is well known that the datacenters hosting today's cloud services waste a significant number of cycles on front-end stalls. However, prior work has provided little insights about the source of these front-end stalls and how to address them. This work analyzes the cause of instruction cache misses at a fleet-wide scale and proposes a new compiler-driven software code prefetching strategy to reduce instruction caches misses by 90%.

Original language	English (US)
Article number	9069187
Pages (from-to)	56-63
Number of pages	8
Journal	IEEE Micro
Volume	40
Issue number	3
DOIs	https://doi.org/10.1109/MM.2020.2986212
State	Published - May 1 2020

All Science Journal Classification (ASJC) codes

Software
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/MM.2020.2986212

Cite this

@article{ad77973e72584854af8b97889bda9bc8,

title = "AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers",

abstract = "It is well known that the datacenters hosting today's cloud services waste a significant number of cycles on front-end stalls. However, prior work has provided little insights about the source of these front-end stalls and how to address them. This work analyzes the cause of instruction cache misses at a fleet-wide scale and proposes a new compiler-driven software code prefetching strategy to reduce instruction caches misses by 90\%.",

author = "Nagendra, \{Nayana Prasad\} and Grant Ayers and August, \{David I.\} and Cho, \{Hyoun Kyu\} and Svilen Kanev and Christos Kozyrakis and Trivikram Krishnamurthy and Heiner Litz and Tipp Moseley and Parthasarathy Ranganathan",

note = "Publisher Copyright: {\textcopyright} 1981-2012 IEEE.",

year = "2020",

month = may,

day = "1",

doi = "10.1109/MM.2020.2986212",

language = "English (US)",

volume = "40",

pages = "56--63",

journal = "IEEE Micro",

issn = "0272-1732",

publisher = "IEEE Computer Society",

number = "3",

}

TY - JOUR

T1 - AsmDB

T2 - Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers

AU - Nagendra, Nayana Prasad

AU - Ayers, Grant

AU - August, David I.

AU - Cho, Hyoun Kyu

AU - Kanev, Svilen

AU - Kozyrakis, Christos

AU - Krishnamurthy, Trivikram

AU - Litz, Heiner

AU - Moseley, Tipp

AU - Ranganathan, Parthasarathy

PY - 2020/5/1

Y1 - 2020/5/1

N2 - It is well known that the datacenters hosting today's cloud services waste a significant number of cycles on front-end stalls. However, prior work has provided little insights about the source of these front-end stalls and how to address them. This work analyzes the cause of instruction cache misses at a fleet-wide scale and proposes a new compiler-driven software code prefetching strategy to reduce instruction caches misses by 90%.

AB - It is well known that the datacenters hosting today's cloud services waste a significant number of cycles on front-end stalls. However, prior work has provided little insights about the source of these front-end stalls and how to address them. This work analyzes the cause of instruction cache misses at a fleet-wide scale and proposes a new compiler-driven software code prefetching strategy to reduce instruction caches misses by 90%.

UR - https://www.scopus.com/pages/publications/85083724981

UR - https://www.scopus.com/pages/publications/85083724981#tab=citedBy

U2 - 10.1109/MM.2020.2986212

DO - 10.1109/MM.2020.2986212

M3 - Article

AN - SCOPUS:85083724981

SN - 0272-1732

VL - 40

SP - 56

EP - 63

JO - IEEE Micro

JF - IEEE Micro

IS - 3

M1 - 9069187

ER -

AsmDB: Understanding and Mitigating Front-End Stalls in Warehouse-Scale Computers

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this