Rethinking software runtimes for disaggregated memory
Abstract
Disaggregated memory can address resource provisioning inefficiencies in current datacenters. Multiple software runtimes for disaggregated memory have been proposed in an attempt to make disaggregated memory practical. These systems rely on the virtual memory subsystem to transparently offer disaggregated memory to applications using a local memory abstraction. Unfortunately, using virtual memory for disaggregation has multiple limitations, including high overhead that comes from the use of page faults to identify what data to fetch and cache locally, and high dirty data amplification that comes from the use of page-granularity for tracking changes to the cached data (4KB or higher).
In this paper, we propose a fundamentally new approach to designing software runtimes for disaggregated memory that addresses these limitations. Our main observation is that we can use cache coherence instead of virtual memory for tracking applications’ memory accesses transparently, at cache-line granularity. This simple idea (1) eliminates page faults from the application critical path when accessing remote data, and (2) decouples the application memory access tracking from the virtual memory page size, enabling cache-line granularity dirty data tracking and eviction. Using this observation, we implemented a new software runtime for disaggregated memory that improves average memory access time by 1.7-5X and reduces dirty data amplification by 2-10X, compared to state-of-the-art systems.
People
BibTex
@INPROCEEDINGS{calciu2021rethinking,
isbn = {978-1-4503-8317-2},
doi = {10.1145/3445814.3446713},
year = {2021-04},
booktitle = {ASPLOS 2021: Proceedings of the 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems},
type = {Conference Paper},
author = {Calciu, Irina and Imran, M. Talha and Puddu, Ivan and Kashyap, Sanidhya and Al Maruf, Hasan and Mutlu, Onur and Kolli, Aasheesh},
abstract = {Disaggregated memory can address resource provisioning inefficiencies in current datacenters. Multiple software runtimes for disaggregated memory have been proposed in an attempt to make disaggregated memory practical. These systems rely on the virtual memory subsystem to transparently offer disaggregated memory to applications using a local memory abstraction. Unfortunately, using virtual memory for disaggregation has multiple limitations, including high overhead that comes from the use of page faults to identify what data to fetch and cache locally, and high dirty data amplification that comes from the use of page-granularity for tracking changes to the cached data (4KB or higher).In this paper, we propose a fundamentally new approach to designing software runtimes for disaggregated memory that addresses these limitations. Our main observation is that we can use cache coherence instead of virtual memory for tracking applications' memory accesses transparently, at cache-line granularity. This simple idea (1) eliminates page faults from the application critical path when accessing remote data, and (2) decouples the application memory access tracking from the virtual memory page size, enabling cache-line granularity dirty data tracking and eviction. Using this observation, we implemented a new software runtime for disaggregated memory that improves average memory access time by 1.7-5X and reduces dirty data amplification by 2-10X, compared to state-of-the-art systems.},
keywords = {disaggregated memory; remote memory; cache coherence},
language = {en},
address = {New York, NY},
publisher = {Association for Computing Machinery},
title = {Rethinking software runtimes for disaggregated memory},
PAGES = {79 - 92},
Note = {26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2021) (virtual); Conference Location: Lausanne, Switzerland; Conference Date: April 19-23, 2021; Due to the Coronavirus (COVID-19) the conference was conducted virtually.}
}Research Collection: 20.500.11850/482467