Modularizing and aggregating physical resources in a datacenter depends not only on low-latency networking, but also on software techniques to deliver such capabilities. In the session we will present some practical features and results of our work, as well as discuss implementation details. Among these features are delivering high-performance, transparent, and partially fault tolerant memory aggregation; and reducing the downtime of live migration using post-copy implementations.
We will present and discuss methods of transparently integrating with the MMU at the OS level, without impacting a running VM;
We will then introfuce the Hecatonchire project ( http://www.hecatonchire.com ) which aim at the disaggregation of datacenter resources and enable true utility computing.
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Hecatonchire kvm forum_2012_benoit_hudzia
1. Memory Aggregation For KVM
Hecatonchire Project
Benoit Hudzia; Sr. Researcher; SAP Research Belfast
With the contribution of Aidan Shribman, Roei Tell, Steve Walsh, Peter Izsak
November 2012
Walk : sequential => each thread start reading from 256k / nb threads * threads id and ends when it reach the start of the following threadsBinary split : we split the memory in Nb threads regions . Each threads will then do a binary split walk within each regionRandom walk : each thread will read a page randomly chosen within the overall memory region ( no duplicate)We Are maxing out the 10GbE bandwith with IWARPWe suspect that we do not have enough core to saturate the QDR linkWe have almost no noticeable degradation when we have Threads > CoresSoftIwarp has a significant overhead ( CPU – latency- Memory use)