PREMA: A Predictive Multi-Task Scheduling Algorithm For Preemptible Neural Processing Units

To amortize cost, cloud vendors providing DNN acceleration as a service to end-users employ consolidation and virtualization to share the underlying resources among multiple DNN service requests. This paper makes a case for a "preemptible" neural processing unit (NPU) and a "predictiv...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings - International Symposium on High-Performance Computer Architecture S. 220 - 233
Hauptverfasser: Choi, Yujeong, Rhu, Minsoo
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.02.2020
Schlagworte:
Computer architecture
Google
Graphics processing units
Random access memory
Servers
Task analysis
Throughput
ISSN:2378-203X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To amortize cost, cloud vendors providing DNN acceleration as a service to end-users employ consolidation and virtualization to share the underlying resources among multiple DNN service requests. This paper makes a case for a "preemptible" neural processing unit (NPU) and a "predictive" multi-task scheduler to meet the latency demands of high-priority inference while maintaining high throughput. We evaluate both the mechanisms that enable NPUs to be preemptible and the policies that utilize them to meet scheduling objectives. We show that preemptive NPU multi-tasking can achieve an average 7.8×, 1.4×, and 4.8× improvement in latency, throughput, and SLA satisfaction, respectively.
ISSN:2378-203X
DOI:10.1109/HPCA47549.2020.00027