Ganesh Ananthanarayanan



I am a Researcher in the Mobility and Networking group at Microsoft Research.

I finished my Ph.D. in the AMP Lab at Berkeley advised by Prof. Ion Stoica. My CV.

Email: ga@microsoft.com


Research

I'm interested in systems and networking. My current research interests are in systems for large-scale data analytics (aka "big data").

  • Geo-distributed Analytics: The next generation of big data analytics will no longer be confined to one datacenter but happen across multiple datacenters, edge clusters etc. How would the storage/compute infrastructure to support geo-distributed analytics look like? [SIGCOMM’15]

  • Datacenter Scheduling: I work on scheduling systems for large clusters whose principles are theoretically-founded. These systems are geared towards straggler mitigation, fairness, and approximate (but timely) results. [SIGCOMM’15][OSDI’14][NSDI’14]

  • Multi-resource Management: Modern analytics jobs are highly complex and use multiple resources – CPU, memory, disk, network. Managing these resources involves multi-dimensional packing of jobs to machines. [SIGCOMM'14]

Selected Publications: (Full List)

  • Low Latency Geo-distributed Data Analytics
    Q. Pu, G. Ananthanarayanan, P. Bodik, S. Kandula, A. Akella, P. Bahl, I. Stoica
    ACM SIGCOMM, London, UK, Aug 2015.

  • PACMan: Coordinated Memory Caching for Parallel Jobs
    G. Ananthanarayanan, A. Ghodsi, A. Wang, D. Borthakur, S. Kandula, S. Shenker, I. Stoica
    USENIX NSDI, San Jose, CA, Apr 2012.

  • Disk-Locality in Datacenter Computing Considered Irrelevant
    G. Ananthanarayanan, A. Ghodsi, S. Shenker, I. Stoica
    USENIX HotOS, Napa, CA, May 2011.

Please see my Ph.D. thesis (Dec 2013) and statement (Apr 2013) for my graduate research.

Selected Talks:

  • Effective Straggler Mitigation: Attack of the Clones [pptx] [pdf]
    AMPLab Summer Retreat, May 2012 and Hortonworks Inc., Nov 2012.
  • Aggressive Cloning of Jobs for Effective Straggler Mitigation [pptx] [pdf]
    USENIX HotCloud, Jun 2012.
  • PACMan: Coordinated Memory Caching for Parallel Jobs [pptx] [pdf]
    USENIX NSDI, Apr 2012, Intel ISTC-CC Retreat, Nov 2012, and VMWare Inc., May 2012.
  • Coordinated In-Memory Caching for Data Intensive Clusters [pptx] [pdf]
    Yahoo! Research, Jun 2011 and Huawei Technologies Co. Ltd., Jul 2011.
  • Disk-Locality in Datacenter Computing Considered Irrelevant [pptx] [pdf]
    USENIX HotOS, May 2011 and Berkeley Cloud Seminar, Feb 2011.
  • Scarlett: Coping with Skewed Popularity Content in MapReduce Clusters [pptx] [pdf]
    ACM EuroSys, Apr 2011, Hortonworks Inc., Jul 2011, Cloudera Inc., Apr 2011, and Facebook Inc., Nov 2010.
  • Reining in the Outliers in MapReduce Clusters using Mantri [pptx] [pdf]
    Yahoo! Research, Nov 2010.
  • Blue-Fi: Enhancing Wi-Fi Prediction using Bluetooth Signals [pptx] [pdf]
    ACM MobiSys, Jun 2009.