Paper Info
Title
Optimizing a shared virtual memory system for a heterogeneous CPU-accelerator platform
Abstract
The client computing platform is moving towards a heterogeneous architecture that combines scalar-oriented CPU cores and throughput-oriented accelerator cores. Recognizing that existing programming models for such heterogeneous platforms are still difficult for most programmers, we advocate a shared virtual memory programming model to improve programmability. In this paper, we focus on performance, and demonstrate that users need not sacrifice performance for programmability. We describe our approaches, experiences, and results in optimizing MYO on a heterogeneous platform consisting of a CPU and an Aubrey Isle accelerator. Our efforts involve the whole system software stack including the OS, runtime, and application.
Year
DOI
Venue
2011
10.1145/1945023.1945035
Operating Systems Review
Keywords
Field
DocType
operating systems,heterogeneous platform,throughput-oriented accelerator core,aubrey isle accelerator,client computing platform,heterogeneous cpu-accelerator platform,whole system software,virtual memory system,heterogeneous architecture,heterogeneous platforms,optimizing myo,programming model,cpu core,virtual memory programming model,shared virtual memory,operating system
System software,Client,Architecture,Programming paradigm,Shared memory,Virtual memory,Computer science,Shared virtual memory,Multi-core processor,Operating system,Embedded system
Journal
Volume
Issue
Citations 
45
1
5
PageRank 
References 
Authors
0.49
8
7
Name
Order
Citations
PageRank
Shoumeng Yan1101.36
Xiaocheng Zhou2111.44
Ying Gao391.01
Hu Chen491.01
Gansha Wu51079.06
Sai Luo691.01
Bratin Saha774239.04