New-age Transactional Systems - Not Your Grandpa's OLTP
John Hugg discusses high volume transaction processing applications with high and low frequency profiles, and how VoltDB can be used for that purpose.
The content has been bookmarked!
There was an error bookmarking this content! Please retry.
Posted by Jonathan Allen on Jun 03, 2008
Though people have been asking for it for years, developers still need write recursive directory copying routines themselves, and every one is nearly identical. So why doesn't this simple and useful function exist in the .NET framework? Actually it does, if you reference the Microsoft.VisualBasic assembly.
Reading and writing to ZIP files is another common task for programmers. A bit more complex than copying directories, developers often turn to third-party libraries or command line tools. Needlessly in fact because ZIP libraries have been in the .NET Framework since the beginning. You just have to dig them out of the J# runtime (and hope the libraries are not decommissioned).
Moving on to our third example, developers often have to read flat files in comma-separated (CSV) and fixed-width formats. While seemingly straightforward, minor points like escaping quoted strings are easily overlooked. Slipped into .NET 2.0 is VB's TextFieldParser, a general-purpose flat file parser suitable these and other similar file types.
So should these little gems remain "language specific", or should they be migrated to the core namespaces of the .NET Framework? A minor question now, but one that is sure to become more pressing as the new languages F#, IronRuby, and IronPython come online.
Using Drools? See what you're missing! Get the Power of Drools with the Assurance of Red Hat
Why NoSQL? A primer on Managing the Transition from RDBMS to NoSQL
SOA All-In-One Guide: KPIs & Best Practices, ESB Report
A Guide to Branching and Merging Patterns
18 agile and lean practices for effective software development governance
This is a great question! I would argue that clearly these kinds of cross cutting concerns belong in the core. There is nothing language specific about any of these at all. Can we add the My library to this list? I've never been able to use it, since it's a VB.Net technology, and I am a C# guy. But it has always seemed really useful to me.
It would be better in the code but - Just reference the .dll from your C# project and then use as usual. For more info see my blog neildoesdotnet.blogspot.com.
Reflect on the Me library and most of it is found elsewhere in the framework but is brought together for convenience. For those bits that aren't the code is pretty simple recreate using Reflector.
It's simply silly that these aren't in the core runtime, especially the zip libraries. I personally see no real reason for ANY functionality that is not totally language specific to be found in a language dll.
Some are really language agnostic and should always be there: for example the zip libraries. I know that to some extent, the restriction on the zip stuff is -partly- because of super complex licensing agreements with patented stuff in the background, so it may not be that simple, but still, the default compression API should be much better than it is now (stream only? ugh!)
For other stuff, its a matter of best practice and backward compatibility. Certain features, especially in VB.NET and J#, are only there for backward compatibility, and do not represent the official best practice for certain tasks in other languages, or have a philosophy that is language specific in design (The My deal and half of what is in the VisualBasic assembly, Eval in JScript.Net, etc). Those shouldn't be there.
If the functionality is that useful however, it should be part of the core, in a flexible enough way that follows the .NET framework guideline and best practices, while being careful not to bloat the framework more than it already is, and the language specific stuff can simply be wrappers around it to give each language their distinctive feel (again: the My stuff in VB.NET does exactly that).
Otherwise, languages that are supported, but have to be compatible with something else through their API, not just the language semantics (IronRuby, F#, etc) would bloat the framework with every new one, features would start getting duplicated, etc.
John Hugg discusses high volume transaction processing applications with high and low frequency profiles, and how VoltDB can be used for that purpose.
Kevlin Henney examines code samples to see what can be learned from them starting from the premise that one won’t write great code unless he knows how to read it.
Jason Ayers share the observations he made watching a team of developers collaborating in real time on the same code base, pushing XP, pair programming and continuous integration to their extremes.
Michael Snoyman presents Yesod, a web framework written in Haskell and containing a web server, templating, ORM, libraries (templating, gravatar, etc.).
Richard Kreuter and Kyle Banker on how to avoid classical RDBMS transactional systems by using compensation mechanisms, transactional messaging or transactional procedures.
Attila Szegedi talks about performance tuning Java and Scala programs at Twitter: how to approach GC problems, the importance of asynchronous I/O, when to use MySQL/Cassandra/Redis, and much more.
One category of risk that project teams need to ensure they address is business value failure – delivering a product that fails to provide value for the business investor.
InfoQ spoke to the authors of Software Systems Architecture on a couple of new topics, the System Context viewpoint and Agile, which have been added to the second edition.
4 comments
Watch Thread Reply