home
:
about uctvs
:
contact
:
uctv
:
u of ca
:
browse by:
campuses
|
conferences
|
hosts
|
seminars
|
speakers
|
subjects
Themis: I/O Efficient MapReduce
(
11/25/2013
)
11
minutes
Conference:
What is Big Data at UCSD?
Host:
UC San Diego Office of Research Affairs
Also available:
Video Podcast
Audio Podcast
YouTube
+ share
description:
Themis tackles the inefficiencies inherent in big data processing engines such as Hadoop. Existing systems sacrifice orders of magnitude of performance for the sake of scalability. For example, Hadoop can be used to sort 100TB of data using thousands of machines. Themis makes such problems feasible using only around 50 machines. While other solutions focus on smaller problems that can be solved using in-memory techniques, Themis is designed for larger problems where efficient use of secondary storage is critical for achieving good performance.
more on this subject:
Local Algorithms
Innovation in Search Engines
iRODS the Integrated Rule Oriented Data-management System
see all from Computer Engineering >
host links:
UCSD Office of Research Affairs Web Site
more from this conference:
Opening Remarks - Sandy Brown and Phil Bourne - Big Data at UCSD
Using Twitter Data to Predict Flu Outbreak
Real-Time Streaming Data from Electricity Grids: Distributed Event Detection and Dynamic Modeling
End-to-end Data Solutions for Simulations at Extreme-scale
see all from this conference >