Loading Events

« All Events

  • This event has passed.

CDS Department Seminar: “Improving user-perceived performance of large scale geo-distributed services”

Mon, 10 Dec 18 @ 3:00 PM -- 4:00 PM

SPEAKER            :  Shankaranarayanan P N, AT&T Labs – Research

TITLE                    : Improving user-perceived performance of large scale geo-distributed services

Date/Time              :  December 10, 2018 (Monday) 3:00 PM

Venue                     :  102 CDS Seminar Hall

ABSTRACT

User perceived end-to-end latency of geo-distributed applications (or services) have a significant impact on the revenue of many businesses. The end-to-end latency is impacted by its many constituent components: (i) User to Application server (front-end) latency, (ii) Application and storage server (back-end) latency, and (iii) cloud and network infrastructure latency. Improving the user-perceived performance of large-scale geo-distributed services is challenging, given their complex operating environments involving user-facing web servers, content distribution network (CDN) servers, multi-tiered application servers, storage servers, and the cloud and network infrastructure that hosts these components. While many novel approaches like SPDY, geo-replicated datastores have been developed to improve the performance of these services, many of these solutions are specific to certain layers, and may have different impact on user-perceived performance. A key goal of our research is to address the above challenges in a holistic manner, focusing specifically on improving the end-to-end performance of geo-distributed services. We evaluate our ideas using real-world deployments, under realistic conditions to show their benefits — on commercial cloud deployments, or through extensive real-world application traces.

BIOGRAPHY

Shankar is a Senior Inventive Scientist at AT&T Labs – Research (Shannon Labs) , where he develops cutting-edge technologies that assist deployment, orchestration and management of network services in federated, massive global scale networks. His research interests are in the areas of Computer Networks, Mobile Systems, Distributed Systems and Cloud Computing, with many of his works explicitly bridging these areas of research. Shankar received his Ph.D. in Electrical and Computer Engineering in 2016 from Purdue University, where he was advised by Prof. Sanjay Rao . Prior to his doctoral studies, he was a software developer for three years at D. E. Shaw & Co. , a New York based hedge fund, working on various technological initiatives within the firm.

Host Faculty: Asst. Prof. Yogesh Simmhan

Details

Date:
Mon, 10 Dec 18
Time:
3:00 PM -- 4:00 PM

Venue

Room No: 102 (Seminar Hall of CDS)