Scheduling - A Secret Sauce For Resource Disaggregation

This repo contains the source code for a constraint-based (Gecode) framework describe in here.

• DDC-RA The framework for finding the optimal memory allocation in an extended rack-scale architecture.
• CPU-RA The source code for a NUCA-aware CPU allocator.

Each folder has a Makefile that can be used to compile the source code.

Requirement

You need to install Gecode to be able to use our source code. A quick recepie would be to run the following commands:

sudo apt-get install libgecode-dev
sudo add-apt-repository ppa:rock-core/qt4
sudo apt-get update
sudo apt-get install libqtcore4 libqt4-dev
sudo ldconfig

Note that Gecode also requires QT libraries. For more information, please refer to the Gecode documentation.

Citing our technical report

If you use our framework in any context, please cite our technical report:

@techreport{farshin-ddc-ra,
   author = {Farshin, Alireza and Roozbeh, Amir and Schulte, Christian and Maguire Jr., {Maguire Jr.}, Gerald Q. and Kosti\'{c}, Dejan},
   institution = {KTH, Network Systems Laboratory (NS Lab)},
   institution = {KTH, Software and Computer systems, SCS},
   institution = {Ericsson Research},
   title = {{Scheduling - A Secret Sauce For Resource Disaggregation}},
   keywords = {Disaggregated Data center, Scheduling, Resource Allocation, Constraint Programming, Gecode},
   year={2021},
   abstract = {This technical report describes the design & implementation of a constraint-based framework for scheduling & resource allocation in a disaggregated data center (DDC) where we build logical servers from disaggregated resources. We show that an Service LevelObjective (SLO)-aware constraint-based solver could improve a data center’s resource utilization by finding better solutions based on provided workload characteristics. },
   series = {Technical Report},
   url={http://kth.diva-portal.org/smash/record.jsf?pid=diva2%3A1599234&dswid=-9580},
}