Posts Tagged ‘database’

A local CS reading group

August 9th, 2013 No comments

Paper Cup, a reading group for computer science papers recently started, based about 30 minutes from me I decided to go along to the first meeting to see what it was like.

The paper under discussion was: Dynamo: Amazon’s Highly Available Key-value Store. I don’t know much about databases and and have never written code that uses a key-store, but since the event was hosted by guys at ebay/PayPal I figured there would be somebody in the room who knew what they were talking about.

The idea behind a paper reading group is that everybody agrees to read a paper before the meeting, then turns up at the meeting and discusses it.

The list of authors takes up three lines and their affiliation is simply listed as As a subject matter outsider who probably reads several hundred papers a year my overall impression was that this paper was relatively information free and was more or less a puff-piece for Amazon. On the other hand it currently has 1,562 citations, a lot more than would be expected for a puff-piece paper published in 2007. I was obviously missing something.

Around 10 people showed up, with a handful sounding very knowledgeable and one person working on a new ‘Dynamo like’ implementation. Several replies to my question of what was so good about this paper, that appeared relatively content free to me, gave the reason that they were inspired by it. Wow, very few scientific papers ever inspire anybody.

The group worked its way through the paper and I tried to nod intelligently at the right time. This is one of those papers that requires lots of reading between the lines, an activity that requires lots of background knowledge and hands-on experience (as an outsider I was only reading the surface text).

I asked if one of the reasons this paper was considered to be important was because it described a commercial implementation rather than a research project. Any design team is much more likely to use techniques outlined in a paper describing a working commercial system than techniques operating in some toy academic environment (papers on Cassandra were appearing at about the same time). I’m not sure the relatively young attendees understood the importance of this point.

The take-away interesting snippet of information: Dynamo gives preference to performance over consistency, if a customer’s shopping basket key-value store becomes inconsistent then information on items added to the basket take precedence over items deleted from the basket (a sensible choice for a retailer such as Amazon).

If you live near west London and are interested in discussing CS paper do join the Paper Cup meetup group, the more the merrier.

SQL usage: schema evolution

January 30th, 2011 No comments

My first serious involvement with SQL, about 15 years ago, was writing a parser for the grammar specified in the ISO SQL-92 Standard. One of the things that surprised me about SQL was how little source code was generally available (for testing) and the almost complete lack of any published papers on SQL usage (its always better to find out about where the pot-holes are from other peoples’ experience).

The source code availability surprie is largely answered by the very close coupling between source and data that occurs with SQL; most SQL source is closely tied to a database schema and unless you have a need to process exactly the same kind of data you are unlikely to have any interest having access to the corresponding SQL source. The growth in usage of MySQL means that these days it is much easier to get hold of large amounts of SQL (large is a relative term here, I suspect that there are probably many orders of magnitude fewer lines of SQL in existence than there is of other popular languages).

In my case I was fortunate in that NIST released their SQL validation suite for beta testing just as I started to test my parser (it had taken me a month to get the grammar into a manageable shape).

Published research on SQL usage continues to be thin on the ground and I was pleased to recently discover a paper combining empirical work on SQL usage with another rarely researched topic, declaration usage e.g., variables and types or in this case schema evolution (for instance, changes in the table columns over time).

The researchers only analyzed one database, the 171 releases of the schema used by Wikipedia between April 2003 and November 2007, but they also made their scripts available for download and hopefully the results of applying them to lots of other databases will be published.

Not being an experienced database person I don’t know how representative the Wikipedia figures are; the number of tables increased from 17 to 34 (100% increase) and the number of columns from 100 to 242 (142%). A factor of two increase sounds like a lot but I suspect that all but one these columns occupy a tiny fraction of the 14GB that is the current English Wikipedia.