Reply to comment

Doug Judd successfully submitted a presentation on Hypertable to Berlin Buzzwords. He describes himself as co-founder and CEO of Hypertable, Inc, a company that provides commercial support for Hypertable, a massively scalable, open source database. Doug started the Hypertable open source project in 2007, while working as an Architect at Zvents, and has been actively building the technology ever since. Doug has over a decade of software engineering experience in the area of distributed computing and information retrieval. He joined Inktomi's Web Search division in 1997 where he held both engineering and management positions. During his five year tenure, he designed and developed large-scale distributed systems, including significant pieces of the crawling and indexing software. Doug earned a B.S. in Computer Science from U.C. Santa Barbara in 1992 and holds four patents in search technology.

Doug will be giving a presentation on "Hypertable: The Ultimate Scaling Machine". Hypertable is a high performance, massively scalable, open source database modeled after Google's Bigtable. Started in March of 2007 by two ex-Inktomi search engineers, Hypertable has been deployed in live production settings since 2008. Project sponsors include Baidu (Nasdaq: BIDU), the leading Internet search services provider in China, and
Rediff.com (Nasdaq: REDF) the largest India-owned and operated Web portal.

Hypertable is unique among the prominent open source scalable databases in that it is implemented in C++ as opposed to a higher level language such as Java or Erlang. The decision to use C++ was made with the same guiding principle that has shaped every decision on the project -- build the world's most high performance, massively scalable database. With a highly optimized, scalable database, efficiency gains scale linearly with the system, dramatically reducing hardware requirements and lowering overall
operational costs.

In this presentation, Doug Judd, the original creator of Hypertable will present Hypertable within the context of the other prominent open source scalable database technologies. In the initial part of the presentation he will describe the Bigtable+GFS design upon which Hypertable is based and will compare it with distributed hash table approaches such as Dynamo (Cassandra, Voldemort) and auto-sharding approaches (MongoDB).

In the second part of the presentation, Doug will dive into the performance specific aspects of the Hypertable implementation. He will describe some of the disk/memory/CPU tradeoffs the system makes in order to deliver optimal performance and will present quantitative evidence that demonstrates that Hypertable is the true performance leader among open source Bigtable implementations.