Skip to main content.
Advanced search >
<< Back to previous page Print

<< Thursday, January 17, 2013 >>

Remind me

Tell a friend

Add to my Google calendar (bCal)

Download to my calendar

Bookmark and ShareShare

Dissertation Talk: Scale-Independent Relational Query Processing

Seminar: Departmental | January 17 | 11 a.m.-12 p.m. | 380 Soda Hall

Michael Armbrust, UC Berkeley EECS

Electrical Engineering and Computer Sciences (EECS)

Newly-released web applications often succumb to a “success disaster,” where overloaded database machines result in high response times, destroying a previously good user experience. Unfortunately, the data independence provided by a traditional relational database system, while useful for agile development, only exacerbates the problem by hiding potentially expensive queries under simple declarative expressions. As a result, developers of these applications are increasingly abandoning relational databases in favor of imperative code written against distributed key/value stores, losing the many benefits of data independence in the process. To counter this regressive trend, I developed PIQL, a declarative language that also provides scale independence. PIQL uses novel static analysis techniques coupled with automatic index creation and incremental precomputation to allow the calculation of an upper bound on the number of key/value store operations that will be performed for any query in an application. When placed on top of our scalable distributed storage system, SCADS, these bounds allow PIQL to ensure consistent performance for complex applications, even as they grow to many hundreds of machines.