For many years there was a solution that was not capable of handling the needs of the services it provided.  An urgent request was made to have this system upgraded to a proper long-term solution.  The goals were to cover 3 key areas, scalability, performance and reliability.  A plan was hatched to create this new solution which went well and in about a month a solution with a formal plan was created.

The solution was to use some currently implemented and some new technologies.  The key technologies were Redhat Enterprise, Samba, Apache and GFS with LVM.  Many of these technologies are used for key back-end systems in the environment but it’s not formally acknowledged of their value.  The plan went ahead with the operations team building the solution and begining to plan how the test and go-live would occur.  The work was done and testing began, everything went well once all areas were identified to ensure the least amount of impact for any live changes.  When the first go-live was attempted there were some challenges with performance of one of the components so the go-live was rolled back for further testing.

Meetings commenced and a set of actions were put in to place to help resolve this key issue.  No real solution was found as no leadership was given to ensure a proper test environment was setup to test this new challenge.  Tweaks were done and the old testing was started and passed with flying colours.  Again the go-live was attempted without any success, yet another rollback.  This cycle occured one more time over the following week with again another rollback.  Still the problem existed but no accurate testing was ever done for this area of the system.

In the end the solution was dropped to hack and tweak the existing solution for some short term gains and increase the workload needed to manage the system.  No additional attempts were made to make the new solution work.

What is hard to understand is why a decision of this scale initially was reviewed in fine detail but to go back to the status quo only took a matter of minutes.  There was no pressing need to make a knee jerk reaction.

Some time has passed since this decision was made and some of the factors of why the choice was made are becoming clear.  The technologies at hand were new in some facets but the operations teams showed the core competency they would be able to handle them and grow their knowledge going forward.  This was used an one of the factors to drop the proposed solution.  The support from the vendor was also flagged in a manner which no vendor would typically support, the operation team was asked to work with support to help create the solution more then troubleshoot, you won’t get professional services from a support agreement, that’s just bad business.

The changes to the current live system have not fully been implemented but time will tell how long these changes can last.

GFSv0.2.2 Released!!

After the success of the last release, Cubika Labs is glad to announce that the new version of GFS in now available. GFSv0.2.2 includes new functionalities and some fixes.

Among important features you will find the spring-security integration, this was the feature most requested by GFS users.

Although other functionalities that we consider vital for a complete integration are still missing, this is an initial approach and the basis to continue working in the new versions.

Some included improvements in this version are:

• spring-security Integration (we use stark-security)
• Tabs group (Thank you so much Nicolas Domina)
• actions into datagrid

If you want report a bug / improvement or vote our Backlog features’ you are welcome to do it in our Jira

See our Confluence for technical documentation

See our SVN if you want to try the code

See new Screencast

In order to explain how this features work, we are going to write some code for a dummy todo-list app

Domain Model

1 – Create Project and Install Plugin

gfs@gfs $ grails create-app todo-list
gfs@todo-list $ grails install-plugin \
[path-plugin]/grails-flex-scaffold-0.2.2.zip
#or
gfs@todo-list $ grails install-plugin flex-scaffold

2 – Generate User, Role and Task

#Generets User and Role for spring-security integration
gfs@todo-list $ grails stark-security-install-full
gfs@todo-list $ grails create-domain-class task

3 – Edit User

class User implements UserDetails {

    String name
    String lastName
    String username
    String password
	String email

    static transients = [
            'authorities',
            'accountNonExpired',
            'enabled',
            'credentialsNonExpired',
            'accountNonLocked'
    ]

    static hasMany = [ roles: Role ]

    //Generates TAB "ADM User" that contains User's tab
    static groupName = "ADM User" 

    // Need to eagerly fetch role relations, because they'll be asked for later
    // by Spring Security when the hibernate session is no longer available
    static mapping = {
      roles lazy:false

    }

    static constraints = {
      name(blank:false)
      lastName(blank:false)
	  email(email:true)
      username(blank:false)
      //Generate 2 textinput to validate password.
      password(widget:"password")
      roles(inPlace:false)
    }

    GrantedAuthority[] getAuthorities() {
        return roles as GrantedAuthority[]
    }

    boolean isAccountNonLocked() {
        return true
    }
    def setAccountNonLocked(boolean nonLocked) {}

    boolean isCredentialsNonExpired() {
        return true
    }
    def setCredentialsNonExpired(boolean nonExpired) {}

    boolean isAccountNonExpired() {
        return true
    }
    def setAccountNonExpired(boolean acctNonExpired) {}

    boolean isEnabled() {
        return true
    }
    def setEnabled(boolean enabled) {}
}

4 – Edit Role

class Role implements GrantedAuthority {

    String description
    String authority

    static final ANONYMOUS = 'IS_AUTHENTICATED_ANONYMOUSLY'
    // Add your roles here so you can reference them, for instance:
    // static final ADMIN_USER = 'ROLE_ADMIN_USER'

    // This list holds all roles, convenient when you're declaring controller methods
    // that should be available to everybody (see AccessController for instance).  When
    // you add roles to your system, make sure you add them to this list as well.
    static final ALL_ROLES = [  ANONYMOUS ]

    //Generates TAB "ADM User" that contains Role's tab
    static groupName = "ADM User"

    int compareTo(Object o) {
        if (o instanceof Role) {
            return this.authority.compareTo(o.authority)
        }
         return 0
    }

    String toString() {
        return authority
    }
}

5 – Edit Task

class Task {

	String name
	String description
	Long priority

	//multiselection:true - A Checkbox appears in all rows on the first datagrid column
	//actions:["postpone"] - show an action button (label=postpone) in each row so you can
	//add bussines logic into postpone method at TaskService
	//You could declare as many actions buttons as you need eg: actions:["postpone","complete"]
	static action = [datagrid:[multiselection:true,actions:["postpone"]]]

    static constraints = {
		name(blank:false)
		priority(widget:"hslider", range:1..10)
		description(minSize:5, maxSize:255)
    }
}

6 – Add to Config.groovy

gfs.security = true

7 – Edit StarkSecurityConfig.groovy and change:

    authorizations = [
        '/': Role.ALL_ROLES,
        '/js/**': Role.ALL_ROLES,
        '/css/**': Role.ALL_ROLES,
        '/images/**': Role.ALL_ROLES,
        '/j_spring_security_logout': Role.ALL_ROLES
    ]

to:

    authorizations = [
        '/**': Role.ALL_ROLES,
        '/js/**': Role.ALL_ROLES,
        '/css/**': Role.ALL_ROLES,
        '/images/**': Role.ALL_ROLES,
        '/j_spring_security_logout': Role.ALL_ROLES
    ]

8 – Edit BootStrap.groovy as below

    import org.codehaus.groovy.grails.plugins.starksecurity.PasswordEncoder

	class BootStrap {

	 def init = { servletContext ->

	  if (!User.findByUsername("admin")) {

      def superUser = new User(username: 'admin',
		           password: PasswordEncoder.encode('admin', 'SHA-256', true),name:"admin",lastName:"admin",
																			email:"admin@admin.com")

	   def role = new Role(authority:"ROLE_USER",description:"Role users")													

	   superUser.addToRoles(role)

	   superUser.save()
	 }

	}

	def destroy = {
	}
   }

9 – Generate Flex Artifacts

gfs@todo-list $ grails generate-all-flex role
gfs@todo-list $ grails generate-all-flex user
gfs@todo-list $ grails generate-all-flex task

10 – Compile Flex and run app

gfs@todo-list $ grails flex-tasks
gfs@todo-list $ grails run-app

11 – Access to application

Tips

Before I start, I need to give a most triumphant prelude. Ladies and Gentlemen… Joe Satriani.

Although there are many different musical instruments, one instrument that truly strikes me as being an instrument for the common people is the guitar. It seems like everyone has a guitar now, and cheap deals for decent guitars are probably the best they’ve ever been. On the opposite end of the spectrum, high end guitars are better than ever as well.

Because the amount of guitarists is so large, just being the best player isn’t the main goal anymore. Several artists have guitars that are “works of art” and are uniquely distinctive. When you see these guitars, you know who plays it. Several people even copy these designs as a tribute, like how I striped my bass as a tribute to Eddie Van Halen. Here are some examples of guitars that are truly iconic:

Eddie Van Halen’s Frankenstrat:

Tom Morello’s Mongrelcaster:

B.B. King’s Lucille:

Bo Diddley’s Big B:

Randy Rhoads’ V:

SRV’s First Wife:

Joe Satriani’s Chrome Boy:

Steve Vai’s Jem:

Prince’s Purple Cloud:

Jimi Hendrix’s Upside-down Strat:

George Harrison’s Rocky:

Buckethead’s Killswitch equipped Les Paul:

Brian May’s Red Special:

Kurt Cobain’s Jag-Stang:

Willie Nelson’s “Trigger”:

I know I left some off. This isn’t because I don’t think they’re iconic, it’s just that sometimes you can only handle so much rockin’ guitars at once. Anyhow, to add to this list is…

Shammy(short for shamrock) Version 3.0. aka: The Jimmycaster!

Here are some of the specs:

Body: Stock Ibanez Stagestar, Agathis(similar to mahogany) in “Shamrock Green”

Neck: Maple, with standard vintage height frets

Fingerboard: Maple

Nut: Brass-fitted by Kittle’s Music

Tuning Heads: Planet Waves Auto Trim Locking Tuners

Pickguard: Warmoth Purple Pearloid, custom cut with help from Jared’s Pickguards and Kittle’s Music

Pickups: 2 Single Coil Stock Ibanez Powersounds with chrome covers, and one GFS VEH(EVH Clone) Humbucker with a nickel cover

Bridge: Stock Ibanez six screw vintage tremolo.  Whammy bar is removed, with the bridge screwed down as close to hardtail as possible.

Straplocks: Dunlop Dual Design

Other features: Chrome Tele knobs, shielded pickup cavity, and Morello Style Toggle Killswitch

…and it comes with a super-shiny Levy’s mirror strap!

Anyhow, it feels great to get this done. I’ve heard and have had various interpretations of what the guitar’s appearance evokes in the minds of its observers. The first one I heard is that it looks like a “Strat from Space.”

The second one, which I should have thought about more, is that it looks like Baby Bop from Barney and Friends

The third one is that it’s a Mardi Gras guitar. It has green, purple, and the maple neck is the gold. The chrome just ties stuff together.

The fourth one is that it looks like a pimp’s guitar. It has the green which symbolizes the money, the maple neck gold which symbolize the honey, the purple which is the color of royalty, and the chrome is there so it is “blingin’.”

The fifth one is my favorite. Often times vintage Stratocasters:

Are compared to vintage automobiles:

while my imported strat copy:

kinda reminds people of a tricked out import tuner:

which is flippin’ sweet!

Anyhow, the main inspiration of this guitar was Eddie Van Halen’s frankenstrat. I know Eddie has issues, and probably isn’t a very pleasant person, but I’ve always loved his tone, his energy, his speed, and his spirit in his playing. I also really like how the frankenstrat was a really cheap guitar that he messed around with. I like that through his messing around, he was instrumental in the creation of the superstrat guitar, and that he was able to make such an iconic guitar out of a cheapy.

I know I kinda ripped off his design a little with my Ibanez bass, eventually I’ll take that tape off and mod my bass’ appearance even further, making a more original design.  I gotta respect Eddie, and that the stripes were his idea, and not mine.

-Ok, enough about the bass. Back to the guitar. I decided not to base my guitars’ appearance on Eddie’s, but I decided I could do some things that were similar to his, while also doing some things that I liked. First of all, it had a maple fingerboard like Eddie’s, and a humbucking pickup in the bridge like Eddie’s(although it sounded nasal as heck, and I definitely wanted to get a nicer one). It also is a strat shape.

Those things were a start, but I had some differences from Eddie’s taste. Now I admit I don’t really like “beat up” guitars. I actually like really shiny and colorful guitars that visually pop on stage. I also like pickguards on a strat, just not in white or black. I like to see them in different colors. Another thing I always wondered was why strats never have chrome pickup covers. Les Pauls usually do, and I thought it would be cool to see on a strat.

So that lead to some thinking… and that thinking went back to some ideas I hadn’t thought about in a long time. Ever since I saw the first Mighty Ducks movie, I liked the color combination of purple and green on their uniforms. On D2, they switched to an eggplant and teal which wasn’t bad either, but I really liked purple and green. They’re kinda opposite colors, but they really pop in a way that doesn’t hurt your eyes like red and blue or red and green sometimes do. Also I realized that very few people have purple guitars, green guitars, and people especially don’t have guitars that are both purple and green.

I decided to go with this color scheme, and I was planning to switch my humbucking pickup. My single coil ones were actually pretty good, so I kept them. After looking around for pickups, I discovered GFS Pickups, which are sold at guitarfetish.com. GFS pickups are on par with Duncan Designed pickups. Yes Seymour Duncan and Dimarzio are American made, but these GFS pickups(like Duncan Designed ones) are way better than most stock pickups, and are sold for very reasonable prices. There really is not that much of a difference in quality. They are about the same. The only problem I ran into with this humbucker is that I had to rout a little under the cavity, which made me nervous, but was no big deal. Routing actually was a very cool “no turning back now” moment, that I kinda remember most of all about the whole project. I then found some chrome covers for my single coil pickups, and still am surprised more strats don’t have these. They look pretty sweet.

So next I took my green Ibanez Stagestar strat copy, got rid of the pickguard that had shamrocks drawn on by a sharpie(that faded), and with my friend Jared’s help, cut a new pearloid pickguard out with a sabre saw. This was really challenging, but a learning experience. I think it took about 2 hours. It was very hard to get things exact, and some cuts around the pickups are sloppy when you look up close, but we were able to cover our mistakes pretty good. Rob at Kittle’s music also did a great job of helping clean up the edges. My advice is to cut two templates/stencils, and secure one to the top, and one to the bottom. We only had a stencil on one side, and that was why we didn’t get everything lined up exactly.

Other extras I put on the guitar were a killswitch, brass nut, and locking tuners. The guitar already had some Dunlop straplocks I bought years ago. Nowadays I use either the Grolsch Straplocks or Planet Waves locking straps. Although the Dunlops are nice and have never failed me, the Grolsch ones have worked for me 99% of the time(I was able to catch it), and the Planet Waves locking straps haven’t failed me yet. Both the Grolsch and Planet Waves are just so much cheaper and do the same job.

Anyhow, the killswitch is just an extra. I do like both Buckethead and Tom Morello, and Eddie uses a killswitch effect on You Really Got Me by turning the volume down on one of his pickups on his Ibanez Desrtroyer(Explorer Copy). I’m still learning where I can use it, and finding opportunities to use it, put it’s pretty cool.  I chose the Morello style toggle because they last longer than the Radio Shack push-button ones Buckethead used to use before he got high quality arcade buttons on his new signature guitar.  Also, you can use the killswitch to turn your guitar off inbetween songs.  Sometimes I’ll find the right spot on the volume knob, turn the guitar down all the way after a song, then when the next song starts, I have to find that sweet spot again.  With a toggle killswitch, I just turn it off, and VOILA!  the volume stays the same when I turn it back on.

I will say that the brass nut and the VEH pickup really changes the sound of this guitar. I’m kinda surprised that the VEH isn’t more “balls to the walls,” but it does handle distortion really well, and picks up harmonics very clearly.  We’ll see how Eddie-like it sounds when I buy a phaser in January.  What I thought was a bonus was that it sounds really good when played clean as well. I thought that this pickup would overpower my stock single coils, but it blends with them quite nicely. The brass nut also gives the guitar more jangle and sustain. I’m surprised that this guitar sounds very smooth, and really beautiful when clean, as well as having some nice crunch with distortion.

The best part is that this guitar was my first guitar, and with the improved sound, added features/aesthetics, and better tuning stability, it feels like a part of me, and has exactly what I would want all-around in a guitar. It feels like some of my soul is in it, and that the guitar and me are one, as crazy as that sounds.

Now although this guitar fits my personality perfectly, I still have some use for Blackie. It’s a great backup, and I enjoy the jumbo frets, higher output humbucker that works better for punk and metal than the Jimmycaster, and the fact that it’s lighter than a Pepsi bottle. Blackie is a very capable backup, but there are a lot of things that the Jimmycaster can do that Blackie can’t do, which should happen with a backup. The First Act does have it’s quirks that make it fun to play, and has some strengths over the Jimmycaster, but it humbly sits aside. The Jimmycaster is Joe Perry, and Blackie is Brad Whitford.

Also, speaking of Blackie, I broke down and tried to VH2 Bumblebee it:

But I stopped halfway through, and removed the tape and reflectors. Blackie is Blackie, and although I modded my Stagestar, and will de-Van Halen-ize my bass to change it’s appearance, Blackie for some reason only looks right stock, and all black.

Anyhow, that’s it.  The Padawan has constructed a lightsaber for himself, and is now a Jedi Master.  If you wish to hear this guitar, wait a few weeks and it will be reviewed at http://rocknrollcowboy.wordpress.com .  I will let you know on this blog when it is up, or you could just read this blog too.  Jared’s a good friend of mine.

Also, Expect a bonus post midweek, I’m already almost done writing it.  I’ll be playing guitar until then.  See you later!

-Jimmy

Doing any last minute shopping to grab food for your feast for this Turkey Day on Thursday?  Or perhaps are you going to buy a bunch of food to give to a shelter so they can serve the needy a warm thanksgiving meal?  Whatever the reason for your search for grub, today’s Referral is the place to go for all the fixings you’ll need to fill some stomachs.

Gordon Food Service Marketplace on the south end of Lansing (on Edgewood Blvd.  near I-96 and Cedar St.) has all sorts of good quality food, and it’s the perfect place to look when you need alot of anything!  Plus you don’t have to be a member or anything to walk in and shop.  The Store Manager Mitch will help you find everything you could possibly need for your meal!

Here’s their website and directions to the store: http://gfs.know-where.com/gfs/cgi/selection?mapid=US〈=en&design=default&country=&region_name=&place=48911&localstore=Go&region=&mapx=&mapy=

Have a great Thanksgiving Tomorrow Everyone!

Global Finishing Solutions, GFS, has invested in a new turret press for its sheet metal processes at its Osseo, Wis., facility.
The new turret, a Murata Motorum Servo-Drive press, is 100 percent electrically operated, using approximately 70 percent less energy than other turrets with similar capabilities, according to GFS. The use of servomotors in place of hydraulics eliminates the need for the use of oils or refrigerant coolers.
The company said the new turret doubles operation speeds, provides improved production capabilities and lowers operating costs.

Here are a few pictures taken at the Gordon Food Service Food Show.  Enjoy!

Now is the time for you to be considering your internship hours for the semesters to come and possible carrer choice!  GFS is a great company to work with.  They can open many possible doors for you in the future into food service, sales, and marketing! 

Gordon Food Service is offering an internship for next summer; the position will be decided in December.  It is GFS’s desire to fill this position with an IPFW Hospitality Management student with the internship being done in Fort Wayne.

The posting will be online for seven days, through 11/27/2009

Do not hesitate in getting your application to GFS! 

Applications are only accepted online at:

http://www.gfs.com/en/careers/current-opportunities.page

Clik on the drop-down: Job Category, Internships-All Departments

Nichts als Arbeit. Das gesamte Wochenende über.

Das Bearbeitenvon GFS, Hausaufgaben und Biologiekarteikärtchen für die Arbeit am Montag ist nicht gerade das, was man unter einem schönen Wochenende versteht.
Außerdem wird mir immer wieder neu bewusst, dass ich alleine bin, Single bin. Überall diese blöde Weihnachtsdekoration und das Beisammensein an den Festtagen wird überall angepriesen.
Es ist ja nicht mal annähernd Weihnachten, aber die Adventszeit beginnt.
Deshalb komm ich mir so einsam vor.

Aber gut, was soll man auch erwarten, wenn man die ganze Zeit am Wochenende daheim bleibt, weil man ständig lernen muss.

Für einen Partner hätte ich wahrscheinlich nicht genug Zeit, bzw könnte ihm nicht die Zeit geben, die ihm zusteht.
Nicht das ich die Zeit nicht weiter geben wollte, aber die Schule frisst immer gerne die freie Zeit.

Und dabei wäre ein Kuss ab und an von einem lieben Menschen jetzt gerade das richtige.

Mir fehlt Liebe.

Maya <3

If you’ve been reading the blog regularly (as if you wouldn’t be!), then you know that I promised to give an update on what one of our computer models is forecasting for Thanksgiving day.

First, though, I want to give you a chance for any new readers (in case you’re not regular readers yet) to look at the previous posts:

• Winter Storm for Thanksgiving Day?
• What Thanksgiving Day Storm?

No Big Storm

Here is the current forecast map for Thanksgiving Day (168-hour GFS from 12z November 19):

Remember, the point of this series of blog posts was to show the change in the computer models from whenThanksgiving was day 15 on the model to when Thanksgiving actually arrives; the purpose was not to give specific forecast details for specific locations.

Now that the forecast is just one week away, however, it’s more reasonable to begin to look at the details of the forecast model more carefully–although a lot can still change during a week, so don’t make any plans based on this one computer model.

Forecast Highlights

• Snow showers in the Midwest and parts of the East (western New York to West Virginia)
• Unseasonably cold in the Midwest
• Some rain and mountain snow for the Pacific Northwest
• Dry and pleasant in the Rockies and Southwest
• Mild to the southern Plains

One More Update Early Next Week

I’ll have another update early next week, and then we’ll see what actually happens and see how accurate the forecast model was.

–Paul Yeager

I shall blog a proper post later on this week. Heh! Promise. (:

Anyway, I enjoyed my Saturday morning with the BFFs at Marina Barrage & Starbucks One Fullerton! We celebrated Az’s 20th Birthday too! ILY babe! <3

Now, next up. When am I going to see the complete GFs together for a meet up session? Don’t look at me anymore. I’m tired of having to organize this things all. *shakebutts*

This is my class project where I choose a topic of Global File System. Where the implementation is on virtual machine, Cluster in a box.

These are the steps:-

1. Set up cluster environment.
2. Configuration of Vmware to create clustered environment ( Cluster in the box).
3. Installation of Centos.
4. Configure RedHat Cluster Suite 5.
5. Set up Logical Volume using Cluster Logical Volume Manager.
6. Deploy GFS clusterally.

If you really interested in this one, let me know, I can foward you my report.

Which filesystems have the largest metadata overhead[1]? We can find out using guestfish sparse file support and a short shell script:

#!/bin/sh -

guestfish=fish/guestfish
testimg=/mnt/tmp/test/test.img
size=1G

for fstype in ext2 ext3 ext4 xfs ntfs msdos ; do
    $guestfish <<EOF
      sparse $testimg $size
      run
      sfdiskM /dev/sda ,
      mkfs $fstype /dev/sda1
EOF
    echo -n "$fstype: "
    du $testimg
done

The results:

Notes

[1] Yes, there are shortcomings in the methodology. What this really measures is how many blocks are written by the mkfs program. Even if you believe the figures, this only measures the initial overhead, but you can use the same technique to measure the overhead of storing (eg) lots of small files, or whatever is appropriate for your workload.
[2] Note the difference between ext2 and ext3/4 seems to be entirely down to the size of the journal, which is a kind of metadata overhead, but one that you can easily control.
[3] Needs three small patches to libguestfs to get these filesystems to work.
[4] This seems off the scale — needs further investigation.

As per usual, my birthday posts have been a bit late nowadays. Just want to wish both beloved sister, Pipah and one of my lovely girl friends, Jann a happy belated. May you both be blessed with happiness and prosperity. Hehe.

Pipah: I don’t have a nice photo of us or should i say of her not covering her face. And i think that’s my other sister’s hand making an appearance lol

Jann: Hurry up come home! *sniff*

PS- i love you both! hehe XOXOXO

I know this is freaking late, been so lazy to blog. shouldn’t be so lazy hmmmm. anyway, happy belated to these lovely ladies; Ryda and Beah!

24th Sep: I love you woman! Best of luck in UK, and hope you will find your happiness in life because you deserve it. Thanks for being there for me all this time. I didn’t want you to go that time but you have to do what you got to do right? i will see you next year darls. MWAHHH!!!

30th Sep: And to Beah, may your wishes come true. One already has. This pic was taken during her engagement. Notice the ring? hehe. happy belated as well babe! xoxo

W Pacific color IR 2330Z 10.05.09 (click image for loop)

for a more recent update on Parma (pepeng) and Typhoon Melor, CLICK HERE

Parma Forecast Track JTWC

The former Typhoon Parma forecast track (JTWC discussion below) is getting uncomfortably close to Luzon in the Philippines.  Basically what is happening is that the powerful Super Typhoon Melor (JTWC discussion below) is racing west-northwestward across the Philippines Sea and its substantial flow is taking the drifting Tropical Storm Parma back south from the same direction that it came.   Melor is being affected a little by this binary interaction with Parma and is now forecast to move a little more northwest prior to its shift in direction.  Typhoon Melor gets caught up in a strong southwesterly flow and it shoots to the northeast and begins to get sheared in the process.  That would weaken Melor as it moves up along the western coast of Japan, though the most recent forecast tracks have been consistently running the storm either across Tokyo Bay or just at the entrance to Tokyo Bay.  However, this latest forecast track has the storm swinging farther west(due to Parma) and therefore the northeastern track has Melor streaking inland over the top of Tokyo itself.  It  remains to be seen if that track comes about but Melor is expected to be a minimum typhoon at that time.   I suspect since it had been so formidable and because of its forward momentum, it will probably have a higher storm surge than what might normally be expected with a weakening typhoon that is transitioning to extra-tropical status.  On the other hand, its quick pace would also mean that the threat for extreme rainfall in Japan may be reduced.

Typhoon Melor Forecast Track JTWC

As for Parma, the forecast suggests that the storm does not strengthen as it moves back south or south-southeast because its moving very slowly back over water that is already been churned up.  That would mean that essentially the warm water on the surface has been blown around and cooler water from below has come up to take its place. Its called upwelling and is why a tropical cyclone needs to stay on the move if its going to survive.    Nevertheless, the issue with the Philippines and particularly the area around Manila, is rainfall and just because  the intensity decreases doesn’t mean that heavy rains are not possible.  Remember, Ketsana was just a tropical storm and flooding from it killed nearly 300 people and some 200 more in Southeast Asia.  Now the boys at the Joint Typhoon Warning Center do kick the storm out to the west before it comes back across the Philippines but it may be close enough to cause problems.  In the meantime, the forecast calls for an increase in intensity as it kicks west and that could mean problems down the road for the same parts of Southeast Asia, namely Vietnam, that also had trouble with Ketsana.

Mon 18Z NOGAPS valid 00Z Wed

Mon 12Z GFS valid 00Z Wed

Here is what is troubling.  I only have access to the 12Z run of the GFS and the 18Z run of the NOGAPS.  Previously, it was the GFS that kept on wanting to take Parma back over the Philippines, most notably right over Manila.  Now, Parma is kinda stuck between a ridge to the west and ridge to the east.  Its got nothing to steer it except the little bit of influence south by Melor.  The thinking is that the ridge to the west will be dominant and take Parma west.  The Joint Typhoon Warning Center seems to be discounting the GFS and NOGAPS as indicated by this sentence in their discussion:  “The available numeric guidance are in wide disagreement with the GFS and NOGAPS attempting an eventual turn to the Northeast and the European and EGRR and ECMWF favoring and eventual westerly track into the South China Sea.”  My concern is that the GFS has been consistent in wanting to bring the storm back over the Philippines one way or another. 

Mon 18Z NOGAPS valid 00Z Thu

Mon 12Z GFS valid 00z Thu

Now, the NOGAPS has jumped on board with that thinking, at least for this model run.  This storm has been exteremly difficult to track due to so many variable influences and the fact that its kinda gotten stuck in no man’s land without strong steering guidance.  In my view, it is still beyond human capabilities to make any definitive statement concerning the ultimate fate in relation to Parma.  I think its safe to say that rain will continue to fall over Luzon, whether or not its intense heavy rain over an extended period of time remains to be seen and we’ll just have to wait and see whether the influence of Melor will be sufficient to run the storm all the way back to the Philippines or if the ridge to the west will nose in strong enough and fast enough to take it away.  Either way, it remains to be seen whether, given the proximity of the storm to the Philippines in any scenario and that it currently has decent upper level venting to allow for development of convection, if the storms move over the islands in the counter-clockwise flow cause more flooding problems.  The government of the Philippines seem to be better prepared and seems to be ready for any eventuality. 

WDPN33 PGTW 051500
MSGID/GENADMIN/NAVMARFCSTCEN PEARL HARBOR HI/JTWC//
SUBJ/PROGNOSTIC REASONING FOR TROPICAL STORM 19W (PARMA) WARNING
NR 32//
RMKS/
1. FOR METEOROLOGISTS.
2. 12 HOUR SUMMARY AND ANALYSIS.
    A. DEEP CONVECTION HAS ONCE AGAIN BEGUN TO FLARE OVER THE
CENTER OF TROPICAL STORM (TS) 19W (PARMA). STRONG VENTING TO
THE NORTHEAST HAS HELPED THE SYSTEM RECUPERATE SOME OF ITS
DEEP CONVECTION LOST NEARLY 24 HOURS AGO FROM THE PASSAGE OF
AN UPPER LEVEL SHORTWAVE TROUGH. A VERY STRONG VERTICAL WIND
SHEAR GRADIENT IS LOCATED JUST TO THE NORTH OF THE STORM. THE
STORM HAS MOVED VERY LITTLE OVER THE PAST 12 HOURS AND REMAINS
IN THE LUZON STRAIT.
3.   FORECAST REASONING
    A. THE FORECAST HAS BEEN SLOWED BEYOND TAU 72 IN ORDER TO
REFLECT DECREASED CONFIDENCE THAT THE SYSTEM WILL EXIT
THIS QUASISTATIONARY STATE BEFORE DAY 5. AS IS TYPICAL IN A
WEAK OR AMBIGUOUS STEERING ENVIRONMENT, THE MODEL TRACKERS
ARE HANDLING THE SITUATION POORLY.
    B. THE SYSTEM IS EXPECTED TO LINGER OFF THE NORTHWEST
COAST OF LUZON FOR THE NEXT 5 DAYS. THERE ARE STILL SOME
INDICATIONS AMONG THE MODEL FIELDS THAT THE STORM WILL
EVENTUALLY PULL OUT TO THE WEST NEAR THE END OF THE FORECAST
PERIOD AS RIDGING RE-ESTABLISHES OVER SOUTHERN CHINA, BUT
AGAIN, THERE IS MUCH UNCERTAINTY. INTENSIFICATION IS NOT
LIKELY BEYOND THE NEXT 24 HOURS DUE TO MARGINAL UPPER LEVEL
CONDITIONS AND UPWELLING OF COOL, SUB-SURFACE WATER. HOWEVER,
DURING THE FIRST 24 HOURS INTENSIFICATION IS POSSIBLE DUE TO
ENHANCED POLEWARD OUTFLOW AHEAD OF THE UPPER LEVEL TROUGH
MENTIONED IN PARA 2.A. THE FORECAST REFLECTS A QUASI-
STATIONARY STEERING ENVIRONMENT WITH A SOUTHERLY DRIFT.
    C. IF THE SYSTEM REMAINS QUASISTATIONARY AT DAY 4 AND
BEYOND, FURTHER WEAKENING IS EXPECTED DUE TO DECREASED OCEAN
HEAT CONTENT FROM UPWELLING. INTENSITIES HAVE BEEN DECREASED
IN THE LATER FORECAST PERIOD TO REFLECT THE INCREASED LIKLIHOOD
OF THIS SCENARIO. //

WDPN34 PGTW 051500
MSGID/GENADMIN/NAVMARFCSTCEN PEARL HARBOR HI/JTWC//
SUBJ/PROGNOSTIC REASONING FOR SUPER TYPHOON 20W (MELOR) WARNING NR 26//
RMKS/
1. FOR METEOROLOGISTS.
2. 12 HOUR SUMMARY AND ANALYSIS.
   A. SUPER TYPHOON (STY) 20W (MELOR) IS CURRENTLY A 135 KNOT
SYSTEM,BASED ON A COMPROMISE BETWEEN THE FINAL-T (6.0) AND
CURRENT INTENSITY (7.0) FROM THE PGTW DVORAK FIX. THE SYSTEM
HAS TRACKED FROM WEST-NORTHWESTWARD AT 050600Z TO NORTHWESTWARD
AT 051200Z AND HAS SLOWED BY 2 KNOTS. THE DECREASED TRACK SPEED
AND INCREASED NORTHWARD TURN ARE BOTH PRECURSORS TO RECURVATURE.
UPPER LEVEL TROUGHING NEAR THE CAROLINE ISLANDS IN MICRONESIA
HAS BEGUN TO ENCROACH ON THE EASTERN SIDE OF THE SYSTEM. AS A
RESULT, FAVORABLE VENTING INTO THIS FEATURE HAS CEASED.
UNRELATED, THE CENTRAL DENSE OVERCAST HAS GROWN CONSIDERABLY,
AND THE EYE IS AT ITS LARGEST DIAMETER YET (32NM). MORE
RECENTLY, DEEP CONVECTION HAS BECOME INCREASINGLY DISPLACED TO
THE NORTHWEST AS THE STY NEARS AN UPPER LEVEL TROUGH NEAR THE
EAST COAST OF ASIA.
3. FORECAST REASONING
   A. THE FORECAST PHILOSOPHY HAS NOT SIGNIFICANTLY CHANGED
SINCE THE PREVIOUS PROGNOSTIC REASONING BULLETIN.
   B. STY 20W WILL CONTINUE TO TRACK ALONG THE SOUTHWESTERN
PERIPHERY OF THE SUBTROPICAL STEERING RIDGE BEFORE ENCOUNTERING
STRONG UPPER LEVEL WESTERLIES ASSOCIATED WITH A DEEP UPPER LEVEL
TROUGH NEAR OKINAWA. THE SYSTEM IS POSITIONED TO VENT AHEAD OF
THIS FEATURE AND MAY ACCOUNT FOR SHORT TERM INTENSIFICATION.
HOWEVER, SHORTLY THEREAFTER, THE TYPHOON WILL BEGIN TO WEAKEN
DUE TO EXCESSIVE VERTICAL WIND SHEAR AND DECREASED OCEAN HEAT
CONTENT. THE PREVIOUSLY MENTIONED TROUGH WILL ALSO ERODE THE
STEERING RIDGE AND INITIATE RECURVATURE INTO THE MID-LATITUDES.
INTERACTION WITH A STRONG BAROCLINIC BOUNDARY (NEAR 30N) WILL
PROMPT EXTRATROPICAL TRANSITION, JUST TO THE SOUTH OF TOKYO.
THE FORECAST REMAINS CONSISTENT WITH THE PREVIOUS FORECASTS AND
CALLS FOR PASSAGE OF MELOR AS A TRANSITIONING CYCLONE ALONG THE
EAST COAST OF JAPAN. THE TYPHOON WILL BE FULLY EXTRATROPICAL BY
TAU 72.//