Skip to content


The following source code and data sets are available:

“Machine Learning at the Mobile Edge [TNSM ’22]:

OrderMatters [NSDI ’22]

NIC Bench [PAM’21]

Connection tracking [HPSR ’21]

SLA-aware resource allocator for disaggregated datacenters [technical report ’21]

PacketMill [ASPLOS’21] 

For our work on DDIO [ATC’20] we have the following releases:

For our work on Path Persistence in the Cloud [CCR’20] we have the following releases

  • Our data is publicly available for reproducibility and further analysis at

Cheetah [NSDI ’20]

For Metron [NSDI ’18]:

Datasets related to our GeoPerf [SOCC ’15] and Kurma [SOCC ’18] papers:

  • 2 weeks of latency measurements that were used with GeoPerf (12GB)

  • 16 months of latency measurements across EC2 (430GB)

  • Synchronised traceroute measurements among 10 EC2 DCs (14 + 4 days, 20 GB)

Code releases at RISE:

– dcpmtool: A tool for predicting and producing a deployment plan for distributed controller planes. The tool accounts for required operational reliability and required control plane traffic.

– aquamet: A technology agnostic link quality metric for radio access networks supporting LTE/WiFi control functions operating in line user-centric throughput requirements.

CacheDirector [EuroSys ’19] releases:

  • ​Slice-aware Memory Management: A set of libraries that can be used to measure NUCA characteristics of LLC and exploited to improve the performance of applications.
  • CacheDirector​: A network I/O solution that extends Data Direct I/O (DDIO), by exploiting slice-aware memory management, and sends each packet’s header directly to the correct slice in the LLC.