GIGGAS' NP-Complete Blog
Tuesday, February 3, 2009
Any suggestions? PM me!
I'll be showing off prototypes through this blog.
Thursday, January 29, 2009
ext4 is the new file system used by Linux. The development effort started on June 28th, 2006, when Theodore Ts'o announced the plan for development. ext4 was marked as stable on October 11th, 2008, which ended the development phase and moved into the maintenance phase. Finally, the first Linux kernel to contain the file system was released on December 25th, 2008.
1. Key Objectives
ext4 started as a series of improvements to the ext3 file systems, including performance enhancements and the addition of 64-bit storage limits. The ext line of file systems are extended file systems mostly associated with the Linux kernel and its many derivative distributions. According to Card, Ts'o, and Tweedie, the main architects of ext2, the “Second Extended File System (ext2) is probably the most widely used filesystem in the Linux community” (1). The improvements that ext3 made to ext2 will be dwarfed by those made by ext4, as ext3 only added journaling in order to increase speed. ext4 increases system/file sizes, along with other major improvements.
2. Difference From Existing Development Efforts
ext4's development effort was different than most due to the simple fact that it is open-source. That means that it is free to access and modify the source code, as long as you upload your changes back into the code tree. Also, this introduced more changes into the line of ext file systems than were incorporated into ext3.
3. Related Costs
As the development effort is a community-based open-source one, the only costs associated with the effort were of development hours. There are no given statistics on how many volunteers gave of their time to code nor of the time they spent developing the new file system. However, it is assured that many talented developers spent thousands of hours working on the ext4 file system during the 2+ years that it was in active development.
4. Consumers Reactions
Based on reactions from the official forums of the Linux distribution, Ubuntu, some users seem to be apprehensive about the prospective speed increases that ext4 is promising. A user named Cypher states: “For an average user of Ubuntu upgrading from EXT3 to EXT4 will yield no real benefit and is probably not really worth the hassle..”, while a user named Frak gives another viewpoint: “...There would probably be no improvement using EXT4. The bottleneck on regular HDDs prevents EXT4 from showing much improvement. You would need to have an SSD [solid state drive] to do anything much noticeable.” However, other users seem to be excited for ext4's speed, such as odinfromvalhalla: “Not sure about if it's ext4's fault but with Jaunty my notebook runs way better than before. I get to book login screen in ~18s (CPU T7500 @2.2Ghz and 4Gb Corsair).” Therefore, the reactions thus far have been varied. It is important to note that even though ext4 is finished with the major development effort, the file system will not be included by default in Ubuntu until the release of Ubuntu 9.04, scheduled for April 2009.
5. Future
ext4's future is bright, as it has not been included into the Linux kernel until recently. As more distributions include ext4 as the default file system, more and more users will be acquainted with it, and that will bring popularity and maintenance requests, which will increase the quality of the system.
References:
1. http://e2fsprogs.sourceforge.net/ext2intro.html
2. http://www.ibm.com/developerworks/linux/library/l-ext4/index.html
3. http://www.redhat.com/support/wpapers/redhat/ext3/#advantages
4. http://kernelnewbies.org/Ext4
5. http://ubuntuforums.org/showthread.php?t=1040332&highlight=ext4&page=2
6. http://ubuntuforums.org/showthread.php?t=1044487&highlight=ext4
Wednesday, January 28, 2009
Monday/Wednesday/Friday:
Linear Algebra and Matrices: We started this semester off with a bang on the first day and studied... FRACTIONS! Luckily we seem to be getting back into math with vectors and dot products, but I know all this stuff and it's aggravating to sit in a class I know I'm probably not going to be challenged at for a while. I mean, I took Calc IV last semester, and this is a major step back.
Senior Seminar (MONDAY ONLY): This class is just a class where we fix up our resumes and get ourselves jobs, and to know what to expect when we get there. We've only met once, and it was ok. Teacher's kind of funny, in a weird kind of way. Got some friends in it so it'll be fine.
Knowledge and Reality: This class is pretty fun. I sit back and let his lecture wash over me, and it's some interesting stuff. He's pretty crazy, so that's always a plus.
Tuesday/Thursday:
Operating Systems: This class is hard to figure out. I had some trouble in the prereq (Computer Organization) because it dealt a lot with hardware (which I only care about if it doesn't work). This class, though, doesn't seem like we're going to be working too terribly much on hardware, mostly working on how OSes deal with hardware. I'm reserving judgement, but I think I'll like it because of the subject matter and the CRAZY teacher.
Software Engineering: I'm really not looking forward to this class. We're working in teams (gah), we're not going to code until the last few weeks of class (mega gah), and we have to write documentation (uber gah!). I know that it's going to be helpful for me to have in the workforce, but I'm going to absolutely hate almost all of the class.
Data Structures and Algorithms: This will be my fun class this semester. I've always been pretty interested in Algorithms. We're looking at efficiency right now, and I never knew Big Oh notation like I do now.
Well, there they all are. If anyone read all of that, thank you. I'm a bit shocked you wanted to read all that, honestly. Oh well.
Thursday, December 4, 2008
^ My most recent songs I've listened to.
^ My most-listened tracks since I signed up with last.fm.
^ Ditto, but with artists.
^ That's specific statistics about my current computer. It runs the latest verson of Kubuntu. Dunno why it's reporting as Debianlenny currently, though.
^ Mah Halo stats. This is important business.
Wednesday, November 19, 2008
Ada
C
Java
PHP
Perl
XML
Bash
HTML
Scheme
Prolog
SQL
PL/SQL
XSLT
Python
C++
Seedings:
1. C
2. PHP
3. PL/SQL
4. XSLT
5. C++
6. XML
7. Python
8. Prolog
9. Bash
10. Ada
11. SQL
12. Java
13. Perl
14. Scheme
15. HTML
First Round Matchups:
#1 C BYE
- C passes through to round 2
#2 PHP vs #15 HTML
- HTML is hardly a language, and PHP is dynamic... easy, PHP wins
#3 PL/SQL vs #14 Scheme
- For whatever reason, I never liked PL/SQL in my Database class... however, I love Scheme in my Organization of Programming Languages class... in a no brainer, Scheme wins
#4 XSLT vs #13 Perl
- The first real head-scratcher. Perl is powerful with strings, while XSLT takes XML and uses it as a database. I think Perl takes it, for me.
#5 C++ vs #12 Java
- This is quite interesting, as Java is supposed to improve off of C++'s mistakes. I think it does, although there are some issues with Java. I'm also far more familiar with Java than C++, Java heads to round 2!
#6 XML vs #11 SQL
- Neither of these are truly programming languages: XML is a markup language where you can determine the tag set, and SQL is just a way to query a database. I use SQL far more often, so SQL it is!
#7 Python vs #10 Ada
- My most fun class at this university was in Ada. It wins hands down.
#8 Prolog vs #9 Bash
- Shell scripting is very powerful, but Prolog is a brain-smashing language that is very very curious to me. I'm being drawn to it like moths to a flame. Prolog.
Round Two Matchups:
#1 C vs #8 Prolog
- C is comfortable and easy, where Prolog is challenging to understand and interesting. The class I had C in was far too easy, and I hardly learned much, while in the class where I have Prolog, there's been much learning. Prolog wins.
#2 PHP vs #10 Ada
- I use PHP far more often than Ada, and I don't like how verbose Ada can get. PHP wins.
#14 Scheme vs #11 SQL
- Scheme kills SQL, as it is so programmer-friendly (to me, at least), and it is a true programming language.
#13 Perl vs #12 Java
- Again, Java is my most comfortable language, and the one I end up heading back to most often. Perl may be powerful with strings, but Java is more familiar, therefore Java wins.
Semifinal Matchups:
#8 Prolog vs #12 Java
- Java is the safe choice, and falls in the Object-Oriented paradigm that is being beaten to death now. Prolog is different, in the declarative mindset, that could give a lot of programmers a giant mindache. In a close one, Java wins. Prolog could make up this ground by winter break, but as of this writing, Java takes it.
#14 Scheme vs #2 PHP
- Scheme is very enjoyable to program for me. I've used PHP a lot, but it is dragged down by the fact that in order to have it look decent on a webpage, you have to shove HTML into it. Scheme wins.
Finals:
#12 Java vs #14 Scheme
- This is the hardest choice I have to make. Scheme is the flavor of the month for me, while Java is the old reliable. The winner is...
#12 Java!
Thank you for following this. It was kind of fun, actually, and I might do it again or something. Woo!
Sunday, November 16, 2008
Fast forward to now.
After today's loss to the Titans, the Jaguars are sitting at 4-6. What happened?
In my opinion, three major things are adding together to end up where the Jaguars are now. Offensive line injuries, opponents being able to stop the run, and Gregg Williams.
Now, of course, the first item leads directly into the second. Losing Manuwai, Mo Williams, Brad Meester, and Collier (whose predicament is a horrible testament to the human race) can lead into a tough time running the ball, and of course, as shown by the Steelers and Patriots in last year's playoffs, teams are now able to figure out how to stop the dynamic duo of Fred Taylor and Maurice Jones-Drew.
Now, Gregg Williams isn't the problem. It's just the same amount of hard-headedness that has been a problem at the University of Michigan. Rich Rod, instead of gradually phasing out Lloyd Carr's style of play over time, has decided to force the spread offense onto the players at Ann Arbor.
In my opinion, that's the same thing that Gregg Williams is doing to the players at Jacksonville. Jacksonville's been forced to adopt the attacking, blitzing defense that Gregg Williams has always favored, and they're not built for that. They're built for Mike Smith's system, which has gone to Atlanta.
By the way, how fantastic is Atlanta? They really turned it around. And a lot of the praise has to be on Mike Smith.
Too bad he's not in Jacksonville.
I don't have too much hope for the rest of the season, but perhaps in the offseason Jacksonville can improve the depth of the offensive line, and go find players more suited for Gregg Williams' style of defense.
Hopefully.
Monday, November 10, 2008
First, Virginia Tech lost to East Carolina. I went from being incredibly hyped up about Tyrod Taylor to watching Sean Glennon do what he loves to do. Lose games. I know Frank Beamer, he's just a kid and he shouldn't be criticized. Maybe you should talk to Mike Gundy about that. He is 40.
After the game, Tyrod's redshirt was burned to the ground. When I heard the news on the radio, I was ecstatic. And all was good until October 25th. In a tough game, Tech went up to Chestnut Hill, Massachusetts.
I watched the whole game. Apparently I'm a glutton for punishment, because the game wasn't even fun to watch. The Hokies offense was horrid all game, and the defense let BC through just enough to knock off the Hokies. All that work, getting back into the top 25, ended in one quick game.
Then Florida State happened. I didn't even see this game, as I was busy getting my *** kicked by VT computer science majors at the 2008 ACM Mid Atlantic Regional Programming Contest. But when I returned home to watch the last part of the game, I wasn't greeted with Tyrod or Sean as the QB. I was met with Mr. Cory Holt as the QB for the Virginia Tech Hokies. As soon as I saw that, my heart sank. Surprisingly, he didn't look bad when I saw him, throwing yet another TD to a non-WR.
And of course this past Thursday. I had nearly had it with the Hokies, threatening to stop watching if they lost the game to the ranked Terrapins. In fact, when the momentum swung towards Maryland in the second half, that was enough for me to start spouting non-sense: "That's it, Maryland wins". Now, remember, at this point, I believe VT was up 20-3. There was one positive out of the game (other than a Hokie win): Darren Evans. This kid is going to be a BEAST!
Here's a recap in 20 words or less:
Virginia Tech inconsistent.
Hokies offense horrible outside of Evans.
GIGGAS at the brink of insanity.
Any questions?
19,127 Forum Posts
8 Blog Entries
837 News/Blog Comments
18 Reader Score Votes
33 Friends
12 Chalkboard Messages
55 Screenshots
15 Videos
8,534 Arena Visits
The GIGGAS's Blog Categories
The GIGGAS's Xbox 360 Gamercard
More
The GIGGAS's Friends
Recent Visitors
The last 10 visitor(s) to this Arena were:
The GIGGAS's Arena has had 8,534 visits
The GIGGAS's Arena has had 8,534 visits
Working...