Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/30951
Title: TMO: Transparent Memory Offloading in Datacenters
Authors: Weiner, Johannes 
Agarwal, Niket 
Schatzberg, Dan 
Yang, Leon 
Wang, Hao 
Sanouillet, Blaise 
Sharma, Bikash 
Heo, Tejun 
Jain, Mayank 
Tang, Chunqiang 
Skarlatos, Dimitrios 
Major Field of Science: Engineering and Technology
Field Category: Civil Engineering
Keywords: Datacenters;Memory Management;Non-volatile Memory;Operating Systems
Issue Date: 28-Feb-2022
Source: 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2022 Virtual, Online, 28 February - 4 March 2022
Conference: International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS 
Abstract: The unrelenting growth of the memory needs of emerging datacenter applications, along with ever increasing cost and volatility of DRAM prices, has led to DRAM being a major infrastructure expense. Alternative technologies, such as NVMe SSDs and upcoming NVM devices, offer higher capacity than DRAM at a fraction of the cost and power. One promising approach is to transparently offload colder memory to cheaper memory technologies via kernel or hypervisor techniques. The key challenge, however, is to develop a datacenter-scale solution that is robust in dealing with diverse workloads and large performance variance of different offload devices such as compressed memory, SSD, and NVM. This paper presents TMO, Meta's transparent memory offloading solution for heterogeneous datacenter environments. TMO introduces a new Linux kernel mechanism that directly measures in realtime the lost work due to resource shortage across CPU, memory, and I/O. Guided by this information and without any prior application knowledge, TMO automatically adjusts how much memory to offload to heterogeneous devices (e.g., compressed memory or SSD) according to the device's performance characteristics and the application's sensitivity to memory-Access slowdown. TMO holistically identifies offloading opportunities from not only the application containers but also the sidecar containers that provide infrastructure-level functions. To maximize memory savings, TMO targets both anonymous memory and file cache, and balances the swap-in rate of anonymous memory and the reload rate of file pages that were recently evicted from the file cache. TMO has been running in production for more than a year, and has saved between 20-32% of the total memory across millions of servers in our large datacenter fleet. We have successfully upstreamed TMO into the Linux kernel.
URI: https://hdl.handle.net/20.500.14279/30951
ISBN: 9781450392051
DOI: 10.1145/3503222.3507731
Rights: © Owner/Author
Attribution-NonCommercial-NoDerivatives 4.0 International
Type: Conference Papers
Affiliation : Meta Inc 
Carnegie Mellon University 
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation

CORE Recommender
Show full item record

SCOPUSTM   
Citations 20

44
checked on Mar 14, 2024

Page view(s) 20

75
Last Week
0
Last month
4
checked on Jun 23, 2024

Google ScholarTM

Check

Altmetric


This item is licensed under a Creative Commons License Creative Commons