ABSTRACT
Scalable analysis on large data sets has been core to the functions of a number of teams at Facebook - both engineering and non-engineering. Apart from ad hoc analysis of data and creation of business intelligence dashboards by analysts across the company, a number of Facebook's site features are also based on analyzing large data sets. These features range from simple reporting applications like Insights for the Facebook Advertisers, to more advanced kinds such as friend recommendations. In order to support this diversity of use cases on the ever increasing amount of data, a flexible infrastructure that scales up in a cost effective manner, is critical. We have leveraged, authored and contributed to a number of open source technologies in order to address these requirements at Facebook. These include Scribe, Hadoop and Hive which together form the cornerstones of the log collection, storage and analytics infrastructure at Facebook. In this paper we will present how these systems have come together and enabled us to implement a data warehouse that stores more than 15PB of data (2.5PB after compression) and loads more than 60TB of new data (10TB after compression) every day. We discuss the motivations behind our design choices, the capabilities of this solution, the challenges that we face in day today operations and future capabilities and improvements that we are working on.
- Apache Hadoop wiki. Available at http://wiki.apache.org/hadoop.Google Scholar
- Apache Hadoop Hive wiki. Available at http://wiki.apache.org/hadoop/Hive.Google Scholar
- Scribe wiki. Available at http://wiki.github.com/facebook/scribe.Google Scholar
- Ghemawat, S., Gobioff, H. and Leung, S. 2003. The Google File System. In Proceedings of the 19th ACM Symposium on Operating Systems Principles (Lake George, NY, Oct. 2003). Google ScholarDigital Library
- Dean, J. and Ghemawat S. 2004. MapReduce: Simplified Data Processing on Large Clusters. In Proceedings of the 6th Symposium on Operating System Design and Implementation (San Francisco, CA, Dec. 2004). OSDI'04. Google ScholarDigital Library
- HDFS Architecture. Available at http://hadoop.apache.org/common/docs/current/hdfs_design.pdf.Google Scholar
- Hive DDL wiki. Available at http://wiki.apache.org/hadoop/Hive/LanguageManual/DDL.Google Scholar
- Thusoo, A., Murthy, R., Sen Sarma, J., Shao, Z., Jain, N., Chakka, P., Anthony, A., Liu, H., Zhang, N. 2010. Hive - A Petabyte Scale Data Warehouse Using Hadoop. In Proceedings of 26th IEEE International Conference on Data Engineering (Long Beach, California, Mar. 2010). ICDE'10.Google ScholarCross Ref
- Ailamaki, A., DeWitt, D. J., Hill, M. D., Skounakis, M. 2001. Weaving Relations for Cache Performance. In Proceedings of 27th Very Large Data Base Conference (Roma, Italy, 2001). VLDB'01. Google ScholarDigital Library
- Zaharia, M., Borthakur, D., Sen Sarma, J., Elmeleegy, K., Shenker, S., Stoica, I. 2009. Job Scheduling for Multi-User MapReduce Clusters. UC Berkeley Technical Report UCB/EECS-2009-55 (Apr. 2009).Google Scholar
- Fan, B., Tantisiriroj, W., Xiao, Lin, Gibson, G. 2009. DiskReduce: RAID for Data-Intensive Scalable Computing. In Proceedings of 4th Petascale Data Storage Workshop Supercomputing Conference (Portland, Oregon, Nov. 2009). Supercomputing PDSW'09. Google ScholarDigital Library
Index Terms
- Data warehousing and analytics infrastructure at facebook
Recommendations
Apache hadoop goes realtime at Facebook
SIGMOD '11: Proceedings of the 2011 ACM SIGMOD International Conference on Management of dataFacebook recently deployed Facebook Messages, its first ever user-facing application built on the Apache Hadoop platform. Apache HBase is a database-like layer built on Hadoop designed to support billions of messages per day. This paper describes the ...
Evaluating SQL-on-Hadoop for Big Data Warehousing on Not-So-Good Hardware
IDEAS '17: Proceedings of the 21st International Database Engineering & Applications SymposiumBig Data is currently conceptualized as data whose volume, variety or velocity impose significant difficulties in traditional techniques and technologies. Big Data Warehousing is emerging as a new concept for Big Data analytics. In this context, SQL-on-...
Scale-out beyond map-reduce
KDD '13: Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data miningThe amount and variety of data being collected in the enterprise is growing at a staggering pace. The default now is to capture and store any and all data, in anticipation of potential future strategic value, and vast amounts of data are being generated ...
Comments