What is it about?

The next generation of big data analytics will no longer be confined to one datacenter but happen across multiple datacenters, edge clusters etc. How will the infrastructure for geo-distributed analytics look like? We present a system that treats the collection of different DCs and edges as one single site, and shows how data analytics can be performed over such geo-distributed data.

Featured Image

Why is it important?

The growing number of datacenters and the rise in edge computing has made the problem of geo-distributed data analytics all the important, whose importance is only growing now with time. A system that automatically distributes computation across these different sites is central to the promise of low-latency data analytics in the wide-area.



Ganesh Ananthanarayanan

Read the Original

This page is a summary of: Low Latency Geo-distributed Data Analytics, ACM SIGCOMM Computer Communication Review, September 2015, ACM (Association for Computing Machinery),
DOI: 10.1145/2829988.2787505.
You can read the full text:




The following have contributed to this page