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 James Vastbinder on Oct 26, 2007
On October 13th at the University of Illinois Eric Traut first showed a version of Windows code-named Windows 7. In the demonstration Eric allocated 40MB of RAM to the virtual machine and ran the stripped down operating system with 10 active processes taking up only about 33MB of RAM.
The actual kernel of the new Windows version is claimed to be about 4MB in size. Further reductions included for a minimal install:
When looking at statistics like this, speculation is rampant. A common theory is that Microsoft is returning a single code base, but no official announcement has been made to date. At present Microsoft maintains several versions of Windows with separate kernels:
A return to a single kernel code base would seem to be a logical next step. Especially given the work being done in the Server and Tools division on the Phoenix project and a move to a single code generation platform.
What do InfoQ readers think?
Using Drools? See what you're missing! Get the Power of Drools with the Assurance of Red Hat
Improve Java Garbage Collection, Runtime Execution, and JVM visibility with Zing
Why NoSQL? A primer on Managing the Transition from RDBMS to NoSQL
Fair Trade Software Licensing - A Guide to Neo4j Licensing Options
18 agile and lean practices for effective software development governance
if you count each released version of the same codebase, minwin is just going to increase the number of kernels ;-)
While this is certainly interesting, it's interesting in an academic way. Hard to say if this is just the next 'WinFS', fun to talk about, but ultimately unrelated to what gets released.
(Besides, we've see Microsoft try the single codebase thing before; they keep oscillating)
It sounds like that's the idea... the beyond vista windows.
My question is: does this mean that Vista is ultimately destined for the same trash heap as WindowsME and MS-Bob? Or is it more of a Windows95 to Windows98 analogy? (i.e. run out vista to get the new thingy out there, make it stable and performant in the followup).
The other REALLY interesting bit is the code name. Since when do you "Code name" as product name/version? And there's the oxymoron... no UI, but lets call it windows?
What I want to know is why does James mixup versions and codebase in his article? Single code base would apply to a generation across server and desktop (such as a single code base for XP and Server 2k3), not across generations (i.e. Server 2000 and Server 2k8).
Jim
is that a signal of massive deployment of windows in micro devices ? take care java
Jim -
I don't think Vista is headed in that direction. I don't have much insight and am providing my own prognostication here. I think the teams are looking to consolidate engineering efforts at the OS level, but I haven't been able to verify as the teams are not speaking to the public yet. Eric's demo is their first communication.
I was concentrating on the recent common theme to reduce competition among internal development teams in similiar solution sets. Recent MSFT examples would be WCF, Workflow Foundation, MS CRM, and Phoenix.
-james
Silverlight is an example of micro .NET. It's multi-platform and small compared to today's standards....
I don't see .NET as being ready as an OS at the micro device level at this time...
Well, then you should check out the .NET Mircro Framework (msdn2.microsoft.com/en-us/embedded/bb267253.aspx). It has no need for an OS, and can just launch your .NET (C#) code. It can be as small as 250K (with minimal CLR libraries).
I no longer have a handle on how Java/Linux/etc. scales from chip to server farms, but Microsoft giving us developers the opportunity of writing in the same language, using the same tools, heck given some proper factoring of code, even running the same code on either a chip or a multi-core, multi-cpu, multi-server environment, is pretty cool. (hmmm, write once, run everywhere, where did I hear that slogan again?)
On the other topic, I know Microsoft attempted to get back to a single codebase during the Longhorn development, and they failed, having to move back to the Windows XP SP2 codebase. So a good number of core changes were cut. One of which was the rework to be done on the I/O subsystem, Windows is just to dependant on the disk, and it's the disk that ultimately makes Windows slow over time.
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.
7 comments
Watch Thread Reply