The Anti-Blog Blog: Everyday Engineering and Architecture

... that some States are expanding investment in Education.

As a Mac user and devotee, this peaked my interest today and added a little bit of warmth to my soul:

[Maine expands MacBook program to high schools by Macworld.com: Yahoo! Tech ]

Too often these days I see internet "journalists" publishing stories without even the simplest fact checking.

Some are simply innocent mistakes while others are intentional slights. Many are simply lack of good journalistic practices.

The sign of good writers and journalists that simply made innocent mistakes is a retraction and follow-up article.

The symptoms of hacks are either more negative articles or ignoring the blatant errors.

A recent article in the Register by Timothy Prickett Morgan titled "Germans fire up 200 teraflop Juropa2 super" is a good example.

Within the first couple of sentences he uses the phrase, "but beleaguered server maker Sun Microsystems wants everyone to know that the box that was turned on last Friday is comprised of its InfiniBand switches and its Xeon blade servers" setting a specific tone for the rest of the article.

Mr Morgan goes on to say "The Sun gear at FZJ is the second big deal that Sun has closed for its "Constellation" HPC clusters, which are comprised of Sun's own InfiniBand switches (nicknamed "Magnum") and its x64-based "Galaxy" blade servers." further in the article.

Here's where things unwind a bit. Mr Morgan then immediately contradicts himself straight-away by writing that "Sun's first big HPC deal for the Constellation machines was the "Tsubame" cluster in Japan, where NEC was the prime contractor." and "Sun's second big Constellation deal was the 433.2 teraflops "Ranger" cluster at the University of Texas (with 62,976 cores using quad-core Opterons)."

If the Tsubame cluster in Japan was first and the Ranger cluster in Texas was second, how could the Juropa2 cluster also be 2nd?!?

That would make it at least 3rd if my math and logic skills are correct.

In reality, Sun has significantly more Constellation sales than the 3 mentioned in the article. Even limiting it to the "large" scale Constellation sales, a quick search of the Internet provides me with at least 5:

• Tsubame in Japan
• Ranger in Texas
• KISTI in Korea
• Juropa2 in Germany
• ACCESS in Australia

According to a Sun press release done on the KISTI super computer, the KISTI cluster is the 4th installation which would put Juropa2 perhaps as the 5th or 6th at a minimum.

So herein lies the rub, was this a simple mistake by Mr. Morgan, an intentional slight or lack of good journalism?

What Mr. Morgan does next will tell. (Hopefully any corrected version will have the same wide re-distribution as the original. )

This blog entry was triggered by Neelakanth's on MySQL InnoDB ZFS Best Practices post.

There's always a running debate about "best practices" and I know several people that say, "there are no *best* practices, only good or better practices".

• Separate the mailstore mailbox partitions (data) from the database files (indices)
• Match ZFS recordsize with page size (8 kb for mbox database, 4 kb for database index files and default of 128k for the actual message store file system)
• Disable File Access Time record
• Stay below 80% space utilization
  

Going beyond those things and considering what Neelakanth wrote, the following things are worth investigating at your own risk*:

• Use a Seperate ZFS Intent Log (slog) - Messaging is definitely write intensive and we've seen some improvements by moving the Intent Log off onto separate drives
• Make sure the ZFS Adaptive replacement cache (ARC) is enabled
• Make sure that the ZFS prefetch is enabled
• Consider using an SSD as L2ARC, SLOG or for the messaging database files / indices
• Determine if your storage is ZFS Write Cache friendly or not -- some storage devices / systems that have NVRAM interpret ZFS's command to flush the write cache in different and un-performance friendly ways. If your storage device has battery backed caches, consider turning off the ZFS cache flush

Interesting article on the internet today...

Sounds like Fort Hays State University is looking at newer, more modern ways to get people quality education for an affordable price. Hmmm, and that's bad? Isn't the job of the university to do exactly that? Deal with scarce resources (eg. time, money, talent) and get the most out of it?

In my dealings with Universities and Colleges I see many things. One of the things I see constantly is this very struggle at many of my customers. What are they really in the business of? "Isn't it obvious, it says 'University' right in their name, Dave?"

Ummm, no not really. Because after seeing how many of them are trying to re-invent the wheel time and time again when it comes to IT related things, it's just boggles the mind. Seems like most institutions have lab equipment purchases down to a science (yes that's a pun). And food services working well.

What seems to be missing in most cases is that extra value add that IT can never get around to because most of their time and money are tied up delivering (or re-inventing) basic commodity services.

Some are at least beginning to ask the right questions:

• Why are we providing email for students when everyone's got one (or more already)?
• Why are we running computer labs when our students have computers already?
• Why are doing X ourselves when most corporations outsource that already?

It's very interesting that Universities and Colleges are looking more like their corporate counterparts in many ways these days. Overall it's a good thing I think.

This is another one of those things you look at and say "Why didn't I invent that?!?"

KeyRingThing

Adium 1.4 beta does Twitter and Facebook! Awesome, awesome.

Always looking for things that let me pare down the number of clients I use and make my life simpler.

I was interested in upgrading my telescope recently but this presents some exciting and intriguing opportunities.

Looks like I might be spending some money renting telescope remotely

Here I thought colleges and universities were all about freedom, creativity and education.

From Slashdot: College Threatens to Sue Students Over Email Addresses

It's always amazing to see things like this. Now I admit I put my foot in my mouth from time-to-time but this takes that to a whole new level.

As Bill and Ted once said, "Be excellent to one another"

hehehe... just had to post this one

Dilbert Make Up a Number

In light of all the news regarding Sun, much of it sour or negative, Joerg Moellenkamp has a good post on the HHOSUG blog about Sun's recent quarter financial performance... folks, it just isn't us, if you're out there saying Sun stinks because of the year-to-year quarter performance, you must think that Intel really stinks at a 26% decline quarter-to-quarter... even the "jewel" of the tech industry Microsoft reported a decrease in revenue and a 32% decline in net income quarter over quarter.

Regarding systems sales companies, Dell (server+network) is down 17% (down 16% as a whole), HP (servers+storage eg. excluding printers and ink cartridge sales) down 18%, Sun down 20%, IBM (systems and technology) down 23.5%...

As the title of this post says, it's relative.

I know the invites have been emailed but I thought I would post an invitation here as well...

More details and the registration form at:

https://www.suneventreg.com//cgi-bin/register.pl?EventID=2708

DATE: Tuesday, April 28, 2009

TIME: 9:00 AM Pacific // 12 NOON Eastern

AGENDA (ALL TIMES IN PACIFIC TIME)

9:00 AM To 9:10 AM Welcome, Introductions, & Agenda

9:10 AM To 9:30 AM What's New in Comms R6 Update 2 Jeff Allison, Product Mgr., Messaging Server

9:30 AM To 9:50 AM Mobile Device Update Marc Daniels, Technical Prod Mgr, Comms Suite

9:50 AM To 10:10 AM Comms R6 Best Practices Greg Balmer, Sr. Architect, Deployment Engineering

10:10 AM To 10:30 AM Convergence Update and Best Practices Arindam Chakraborty, Product Mgr., Convergence

10:30 AM To 10:40 AM Roadmap Update Jeff Allison, Product Mgr., Messaging Server

10:40 AM To 10:55 AM User Roundtable

10:55 AM To 11:00 AM Close, Next Steps

Virtual Box is #3 on the list

Sweet!

Managing by Spreadsheet (mBS)

It never ceases to amaze how many of my customers with million dollar budgets still rely upon homegrown spreadsheets to manage their projects. Large or small, doesn't matter -- it's spreadsheet. Oh sometimes it's color coded to indicated status, you know, red for behind schedule, yellow for mostly on schedule but about ready to slip and green if it's lucky enough to be ahead or on schedule (what that purple, brown, orange color coding means I have no clue). About 1/2 the time it's up-to-date, the other 1/2 it's not and in some cases it's really far out of date. Many times it's shared via email and some of the times it's even kept out on the server were team members can download the latest copy. Heaven forbid they should want to change it themselves -- majority of the time they email their changes to the Project Manager or whoever "owns" the spreadsheet... ahem, project plan.

Come on, we all know organizations that do exactly this. Let's not kid ourselves we also do it as well ("Hi I'm Dave. I suffer from mBS syndrome")

The Trouble in Spreadsheet-town

The main issue is that project management by spreadsheet just doesn't do the trick. There's too much data re-entry (I'll send you an update via email and you update the sheet for me) and in most cases things aren't up-to-date. Additionally the data needed isn't tracked to the level needed (eg. the spreadsheets are really high level lest they cause even more data re-entry) and performance metrics really aren't tracked over time so you can't improve forecasting as the project goes along or apply some form of multiplier to the next project very easily.

The Trouble with Microsoft Project

At the other end of the spectrum, you've got Microsoft Project and other very high end tools that are capable of tracking the largest of large projects (like building submarines, sky scrapers, shopping centers and malls). These have too much detail and the tools themselves get in the way of managing the project. There's a genuine and legitimate fear of these types of tools.

Some Simple Options

Enter the web-based tools. There are now many offerings (almost too many in fact) that are web-based and occupy the middle ground between spreadsheets and the Microsoft projects of the world. Even better yet, there are now some specialized tools for managing Agile development work as well (eg. scrums and defects). Awesome!

At the very basic end of things, there's a category of project tools which really have a focus on the collaborative aspect of projects. The emphasis here is on collaboration and not necessarily enforcing good project management discipline or even capturing project metrics. What these tools is get the project plan and general info off someone's hard drive, laptop or home directory and make it a group effort. Keeping it up-to-date is no longer one person's job. Adding new issues or items is no longer one person's job. Everyone can see what's going on and if everyone keeps it up-to-date, then everyone sees the up-to-date data. Neat!
Some eliminate the need for the spreadsheet by providing basic task breakdown, other services out there simply add collaboration to that spreadsheet.

There are quite a few in this category but two I've recently run across that seem to be decent, one called Huddle and one called Basecamp. Think social network meets task / project management. Good basic stuff and better than mBS.

High End

At the high end you find things like Project Insight which, while not quite like Microsoft Project, can actually offer up about 75% to 80% of Microsoft Project's features via the web. Or for those just starting out, Project Insight can start smaller and offer something like 25% to 40% of Microsoft Project's core features in a workgroup edition.

I know. I know. What about Open Source or FLOSS (free as in libre open source software)? Yes Virginia, there are even open source offerings in this high end category. One such service and software package is called Project.net. Very feature rich and high end, yet open source. Yes, I put Project.net in the high end category.

Middle of the Road

Sometimes we can't simply get that $300,000 Italian sports car and must settle for a good old Chevy with a decent stereo system.
More middle of the road web-based project management adds true project management features beyond the collaboration. You're now tracking meaningful metrics such as time spent vs. projected, risk levels and issues, etc. Things that help you not only deliver the current project but also will help you plan / estimate the next one as well.

At the very low end of this category would be QuickBase from Intuit which I consider more a bridge between very basic and middle of the road. Good clean user interface (it's from Intuit, the Quicken people, would you expect otherwise?). No QuickBase isn't for complex projects or project portfolio management (PPM) but it's a decent online project management system with a good UI. I had a customer complain that QuickBase was too simple of a tool. Um, I got news for you it's still light years ahead of doing it by spreadsheets IMHO.

On the high end of the middle of the road (MOTR) category you find things like Daptiv's platform which can do many many things and even has an API you can develop integrated applications against if you are so inclined. Again Daptiv is a hosted solution like QuickBase.
If you're looking for something open sourced web-based either hosted or on-prem in the MOTR category, I like TeamWork -- it's award winning open source which also has an Agile component. Nice.

Ok, I confess I also like Copper for it's UI, is well done, modern and easy to use (can a UI be sexy?). The design reminds me alot of Apple Mac for some reason -- well done and useful perhaps?

One other one that's solidly in this category is Team Dynamix which more than a couple of my Higher Education customers use. Team Dynamix is something you can start simple and grow with but doesn't do much specifically with Agile.

Specifically for Agile

Agile development brings new things to manage in a faster paced environment. (Yes I've seen Agile management via spreadsheet. Again, no no no. Have we not learned anything from the project management by spreadsheet attempts?). Luckily there are some offerings out there to assist with Agile development projects.

I mentioned that TeamWork has an Agile module for managing the scrum, so it can do basic project management and agile scrum management which is pretty good. If you want something that focuses more on just the development aspects of agile projects, the Rally and VersionOne are a couple of options.

Rally kind of seems like the Microsoft Project in this category, really feature rich and often more than what people need so I am not so impressed with it. I'd also consider VersionOne as being at the high end of the agile project management tools here -- it does seem easier or perhaps a bit more user friendly though.

Personally, I also like TargetProcess, ThoughtWorks and AgileBuddy for tools that are crisp, easy to use and really beneficial in this area. ThoughtWorks is really high on the list, AgileBuddy and TargetProcess come in a close 2nd with VersionOne running 3rd in my mind.

Recent Recommendation

Recently I had an opportunity to take a look at one particular customer project. Doing quite a bit of development work and had a real need for testing, solid promote to production cycle management and upgrading their overall project management to something more than spreadsheets.
I basically told them to look at the ThoughtWorks product set of Mingle, Cruise and Twist right away for the development, testing and promotion cycle management. And then review Team Dynamix or Copper for the project management side of things. If I had to choose between Team Dynamix and Copper, my deciding factor would be project portfolio management (PPM) -- if PPM wasn't a requirement, I'd go with Copper.

Have you ever wondered why "thin clients" haven't really captured more of the corporate desktop?

Or "Eco Computing"?

Wondering why that SOA initiative or Identity Management initiative really isn't farther along that it is?

And what's up with all the hype and argument about Cloud Computing? Is it a good thing or isn't it?

Well, Mike Kavis has an excellent post that although it's aimed at Cloud Computing, it really applies to most major compute changes within an organization... it's not about the technology, it's about the people.

A good article over at Slashdot points out "Why IT Won't [or Can't] Power Down PCs"...

It's amazing that all this comes down to governance really... then again I am not really surprise because those organizations that's one of the large differences between organizations that exist at the lower end of the Maturity Model spectrum versus those that manage to break through the ceiling of the 2nd quartile... simply put, governance... the ability to see the larger picture and make effective decisions...