A short introduction to Google Big Table. What is it and
why is it important in the big data field. Why is it important
to consider what Google is doing now and what they plan to do ?
developed by Google • In-house database for very large data sets • Has influenced the NoSql db market place • Highly distributed • Client based validation • Row / column / timestamp indexing • No Joins available • Based on assumption of write once read many www.semtech-solutions.co.nz [email protected]
Google file system and Chubby Lock service • Storage and versioning based upon time stamps – Data sorted by youngest first • No validation, validation left to client • Allows for hardware and system failure via redundancy • Arbitrary row columns • Arbitrary column data types • Very high data throughput www.semtech-solutions.co.nz [email protected]
columns grouped into families – Column families named • Access granted at the family level • Family members grouped in compression • No column datatype constraint • Integrated with MapReduce – I/O data streaming www.semtech-solutions.co.nz [email protected]
leads the way in terms of functionality • Their market needs drive their systems and offerings • What they have done to date will – Affect the rest of the big data market – In the future – Via cross pollination of ideas – Via emulation of their success www.semtech-solutions.co.nz [email protected]
www.semtech-solutions.co.nz – [email protected] • We offer IT project consultancy • We are happy to hear about your problems • You can just pay for those hours that you need • To solve your problems