Documents

Big Data in Enterprise challenges & opportunities.pdf

Description
Big Data in Enterprise challenges & opportunities Yuanhao Sun 孙元浩 yuanhao.sun@intel.com Software and Service Group Big Data Phenomenon 20TB/hour Sensor output of a Boeing jet engine $20+B Acquisitions in the last 12 months 1.8ZB in 2011 2 Days the dawn of civilization to 2003 750M Photos uploaded to Facebook in 2 days 200PB Created by a Smart City project in China Data are becoming the new raw material of business: an economic input almost on a par with capital and l
Categories
Published
of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Big Data in Enterprise challenges & opportunities Yuanhao Sun 浩  yuanhao sun@intel com Software and Service Group  Big Data Phenomenon 20TB  /hour Sensor output of a Boeing jet engine $20+B Acquisitions in the last 12 months 1.8ZB in 2011 2 Days > the dawn of civilization to 2003 750M Photos uploaded to Facebook in 2 days 200PB Created by a Smart City project in China Data are becoming the new raw material of business : an economic input almost on a par with capital and labor. The Economist, 2010 Information will be the “ oil of the 21st century  ”.  Gartner, 2010 966PB Stored in US manufacturing (2009) 200+TB A boy’s 240’000 hours by a MIT Media Lab geek $800B in personal location data within 10 years $300B  /year US healthcare saving from Big Data  Big Data in Telecom ã Lots of data – One telco operator: 360TB Call Data Records within 6months (in a provincial branch, 100M users) – The other operator: ~300TB web access logs from mobile phones within 6 months ã Keep growing – ~2TB CDR/day in a provincial branch ã Various data sources: – CDR (Voice, SMS, GPRS, 3G, WLAN, Value-add services, etc) – Billing & accounting data, sales & marketing data, etc. – Web access logs – Network signaling data – Base station sensor data  0 500 1000 1500 2000 2500 3000 3500 Open Source HBase (0.90.3) Optimized HDFS I/O 700 3500 How Hadoop helps ã Map/Reduce for data loading and data cleansing ã HBase as the data store − Inserting 10000 records/second/server (2-way, 32GB) in average − Read from disk: >400 query/second/server, latency within one second (0.05s~0.8s under different load) ã A query is a scan to get all CDR within one month for one user. ã Optimizations significantly increase the throughput of a 8-node cluster 2011/12/5 4 query/s 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 Open Source HBase (0.90.3) Advanced Region Balancing 25000 82000 insertion/s
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x