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

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

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 m...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE MICRO 2020-05, Vol.40 (3), p.56-63
Hauptverfasser: Nagendra, Nayana Prasad, Ayers, Grant, August, David I., Cho, Hyoun Kyu, Kanev, Svilen, Kozyrakis, Christos, Krishnamurthy, Trivikram, Litz, Heiner, Moseley, Tipp, Ranganathan, Parthasarathy
Format: Artikel
Sprache:eng
Schlagworte:
Cloud computing
Complexity theory
Data centers
Hardware
Optimization
Prefetching
Servers
Warehouses
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung: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%.
ISSN:0272-1732
1937-4143
DOI:10.1109/MM.2020.2986212