Log on:
Powered by Elgg

Marlon Bright :: Blog :: Archives

May 2008

May 16, 2008

We have now gotten to our apartment and are settling in.  We are very tired and it has been quite a journey already.  We knew that we were the first students to take part in this program so it is going to be a learning experience for all of us.  First things first, Spain is expensive! We are going to really have to watch our money! We arrived to check into to our flat to the surprise that we had to pay all expenses up front in cash which totaled up to 1197 Euro which translates into $1987.  Naturally, each of us except for Juan only brought less than 300 Euro cash with them.  So Justin, Javier, and I had to scramble to an ATM to try and withdrawal the rest before we could enter our accommodations.  Javier and I managed to withdraw enough money, but because of transaction limits we were only able to do so through 3 withdrawals from two different bank accounts which means that we incurred withdrawal fees from both the Spanish bank and our banks back in the states which added to the expense.  For future students, I highly recommend that they take note of this and bring this money with them from the states. 

 

Our first flight left Miami at 1:55pm local time on Wednesday, May 14.  I guess we love the adventure because we cut it close many times boarding.  Javier ran a little late getting to the airport, but was fortunate enough that his bags were not loss.  In fact he was one of the first to receive his bags in Barcelona.  Justin stepped outside of security to take a smoke and unfortunately forgot to bring his boarding pass with him, so Juan had to make a dash out of security to reunite Justin with his pass.  Fortunately, we all made it back in reasonable time. 

Our first flight arrived in Philadelphia around 4:40pm local time.  We had quite a walk to change concourses and aircraft for our flight departing for Barcelona at 6:20pm.  Again, Javier loves adventure and set out on a quest for a famous Philly Cheesesteak.  He found it, but boarded the plane on the final boarding call. We were airborne from Philadelphia about 6:30pm local time headed across the Atlantic Ocean for a flight of a little over 8 hours.  It was a very interesting flight due to the fact that we were traveling sort of “into time”.  We experienced a sunset and a sunrise in a matter of about five hours, resulting in a pretty short night.  None of us was able to sleep that much on the flight so we are paying for it now.  Our flight arrived in Barcelona about 8:30am local time the next day.  We were met my Ivan from the Barcelona Supercomputing Center who showed us a great deal of hospitality and for that we are very thankful.  He helped us get a taxi which was quite an experience with luggage for 4 people and 5 people piling in; but, we managed to all fit in tying four pieces of luggage to the top of the van.  He then went with us and helped us work through the ordeal of acquiring our housing.  After that, he accompanied us to our flat, gave us much instruction about the city, and then took us to ensure that we mastered the use of the public transit system.  After that he headed to campus and we to a long awaited breakfast… or I guess it was lunch at 1:30pm, but with the 6 hour time difference you get the picture.  By the way, the Cafeteria Aran next to the Fontana Metro Station is cheap but no good.  The sandwiches were cold and hard and the air in the tiny place was filled with cigarette smoke (no Florida clean indoor air act).  Still everything was not all that bad until halfway through our meal we noticed the attendants serving the next customers…lighted cigarette in hand.

 

  

Keywords: Spain Day 1

Posted by Marlon Bright | 0 comment(s)

May 19, 2008

Today was our first day at the Barcelona Supercomputing Center which is on the campus of UPC.  We were pretty glad to know that it will only take us approximately 25 minutes door to door to get to campus from our flat by train.  The train system here is very good…it sure beats the hustle and bustle of morning gridlock in Miami.  Our first stop on campus was the Ia Oficina de Mobilitat Internacional or translated to English the International Students office to register and receive some welcome material.  After that we were taken to where we would be working with a slight and unintentional detour by a chapel that houses one of the largest computers in the world.  Yes I said a chapel.  Obviously, our guide for the time was not too familiar with where we were bound, but he did manage to unintentionally introduce us to the securely guarded chapel that houses MareNostrum, a supercomputer with 10,000 cpu’s.  I was very interested to hear the story behind this so I inquired.  The basic synopsis was that the building, although it looked like a chapel, had not been a chapel for some time now.  In more recent years it was used as meeting space.  When the BSC was preparing to purchase MareNostrum, it just happened to be the best available and secure place to house the mammoth configuration of computer hardware.

 

After a slight detour, we managed to run into my advisor, Dr. Rosa Badia.  Again, like Ivan Rodero who met as at the airport, she was ever so helpful in getting us acquainted to our new surroundings.  We did our usual organizational check-ins receiving access cards, desk space, and trying to work out the kinks to getting our laptops access to the network.  Around midday Juan, Javier, and I had lunch with Rosa at one of the Campus Cantina’s.  I am not sure of the name of it, but I was very pleased with it.  I had the opportunity to practice my suspect Spanish will ordering, and was able to order a three course lunch for only 6.60 Euro which is roughly $11.  This was a wonderful bang for your buck considering the food prices we have seen so far.  I am sure that I will learn its name very well as I intend to frequent it a lot.

After lunch we met with Dr. Badia and began to layout our plan of research for the summer.  We etched out that our first objectives would be to work with the developers of the profiling and predicting tools Paraver and Dimemas to get them properly functioning on our cluster’s Mind and GCB back at FIU.  This should involve making sure that the trace generating tool, mpidtrace, is properly linked into the WRF code which we are executing for benchmarking purposes.  Once this is done we can compare how our home grown applications, Amon and Aprof, work in comparison to their prediction tool, Dimemas.  I anticipate great strides in this work as we are already somewhat familiar with these tools from working with them in Miami.  Now, we have the opportunity to have all of our questions answered as we work along beside the developers of these applications.  Once this is done we will move to installing Amon and Aprof in MareNostrum and testing its ability to scale to large amounts of nodes.  From here we are sort of open-ended as we have been told that the ask of getting Amon and Aprof up and working on MareNostrum may be a bit complicated.  Ultimately our goal is to improve application performance predictions on clusters to aid in the meta-scheduling process of the grid enablement of scientific applications such as the Weather Research and Forecasting Code (WRF).

After a pretty full day of getting acquainted and setting the foundation for a good bit of work ahead, we rendezvoused back at that flat and debriefed each other on our days.  We are still in debate about getting internet connection in our flat.  It seems quite expensive to get in installed, but I think we may pay the price for the convenience.  We haven’t noticed any internet cafes around, but then again we haven’t fully surveyed our immediate area.  I plan to take a run tomorrow morning and check out what is in our neighborhood.  The other minor issue that we have is that there is four of us and we were only issued three keys…go figure. In addition, the key for the iron gate door to the building is an old fashioned skeleton key so we are on a bit of an adventure to find a key maker that can reproduce this.   Personally, I am in a debate as whether or not to rent a local cell phone.  I am the only one of the guys who does not have one.  It is probably a good idea to have one for convenience and in case of emergency so I will definitely look into to getting one.

 

After a fine dinner of ham and cheese sandwiches and ritz crackers, we decided to hit the town for the first time in Barcelona.  Our flat is in a pretty convenient location so we were in the heart of the city on the train in under ten minutes.  We got off the metro on the famous “Las Ramblas” Avenue and headed the few short blocks to the shore of the Mediterranean Sea.  At the end of the avenue pointing out into the Mediterranean Sea was a fairly large tower topped with a statue of the great explorer “Colom” or as we know him Cristopher Columbus.  We walked around the seaport area and up Las Ramblas for quite some time.  We took some breaks at the epicenter of the city, La Placa de Catalunya, at a quare where a traditional dance, accordion, violin, and all, was taking place, and in a tapas bar for a few refreshments and conversation.  We were even greeted on Las Ramblas by the familiar signs of McDonalds, Burger King, and Kentucky Fried Chicken (Yes Kentucky Fried in Spain.  I wonder if Spaniards even know where Kentucky is?).  However, these establishments looked more like fine dining restaurants that the sometimes less than clean fast food joints that they are in the United States and their prices reflected it.  A double whopper value meal cost 7.20 Euro! That is nearly $12!  This exchange rate is brutal.  I think that I will wait till I get back to the states and pay half of that.  Obviously, the Spain franchise didn’t get the memo about Burger King being the cheap and fast food alternative.  Soon after we wrapped up our night and returned home looking forward to further exploring this weekend.  However, I will be doing good to make it to bed soon.  After trying to stay up most of the day after the short night caused by our travel, I ended up falling to sleep at around 7pm and stayed that was with only one awakening until 8am this morning…I am pretty well rested now.

   

One other thing I noticed today while at the BSC (Barcelona SuperComputing Center) that was a wake-up call to me, but really I should have realized it.  The computers that we registered on were in Spanish.  Also, they had a different keyboard that had some different characters and some usual characters in a different place.  This did make navigation and typing just a bit different as you had to put a little more thought into what you were doing.  Another thing that I realized was that they did not make different versions of programming languages (C, Java, Perl, etc.) in different languages (i.e. Spanish, French, etc.).  The declarations and command for these languages derive from the English language, which makes sense because most of computer technology’s developments were in the United States.  I was told that differences often came that the comments for programs would be written in another language, but this was infrequent as well as it was not conducive to being able to distribute software globally.  With this I am encouraged because, this almost demands that Computer Scientists globally learn to communicate in English.  However, it makes me feel guilty about thinking that my studies were that hard.  I only had to learn the technical aspects, not a totally different language!  We American students are a bit spoiled, but we better get it in gear.

These observations already allow me to see how fortunate I am to have this experience.  Most technical programs do not offer this experience since Computer Technology is heavily focused in the US.  Study Abroad programs for Americans are usually more reserved for Business and Liberal Arts majors; but, it is important for us to realize the global world that we live in and serve with the development of technology and the fact that their needs may differ from ours in some ways and that we should take that into consideration in our work.  This makes me very interested to read about the experiences of my program counterparts in India and China.  Talk about a change in culture!  At least I am somewhat used to everyone around me speaking Spanish (although a different dialect) having lived in Miami now for three years.

Keywords: Barcelona, Day 2, Spain

Posted by Marlon Bright | 0 comment(s)

This morning’s jog was quite refreshing.  It gave me a chance to stretch my legs (even though we have been doing a lot of walking already) and it allowed me to see our neighborhood of Barcelona above ground (as we have been mostly traveling my underground metro).  The first thing I noticed was that there were plenty of people out and walking around, but there weren’t that many people out for jogs, at least in our immediate neighborhood.  They do have a nice tree covered median between the different directions of traffic along the major streets outfitted with paved paths for walkers, joggers, and bikers.  Another thing that made the jog a little different is that a good portion of the neighborhood streets in Barcelona (with exception of the main ones) are narrow, one way, streets…and I thought New York had building on top of building.  It does give the neighborhood a somewhat cozy feel.  And, oh yeah, I found that the golden arches are never far as there is a McDonald’s almost 4 blocks from our flat.

Some interesting tidbits that I have noticed a bout Barcelona so far:

  • The Simpsons are relatively big in Barcelona.  At the street festival today, they had an inflatable jumping pin for children complete with the whole gang of cast members.  Also, on the first day before our TV gave out, I turned it on to see the Simpsons in Spanish.  However, I know that the characters have Spanish accents, but they sound nothing like their American counterparts!
  • I have been told that the tap water here in Barcelona is perfectly safe to drink, but that it has a somewhat funny taste.  I second that statement.  I have never been able to tell such a difference between tap water and bottled.  I guess there is something good to be said about good ole’ Rocky Top (that’s Tennessee where I am from).

This afternoon we headed out for our first full day of exploration on Barcelona.  We began in our own neighborhood at the festival I had discovered while taking a run.  We couldn’t determine whether it was a special festival or just what Barcelonans did on Saturday.  The highlight of that trip for me was the free chocolate bar that they were handing out which was larger than any chocolate bar I have ever had and oh was it good!  From there we did some more exploring, taking a different route to another nearby subway line.  Walking along that way allowed me to take in the fact that there is a bakery on nearly every block.  Spaniards love their fresh baked bread, and the aroma is very pleasing to the nose.  Our next stop was downtown where we took the chance to take in the sites of Port Vell and La Aquarium, Barcelona’s Aquarium.  The odd thing about Port Vell on this evening was the fact that there was a British battleship docked alongside with armed soldiers pacing its decks and a Spanish Flag flying over it (we knew it was a British vessel by a flag on the front of the ship and the accent of the soldiers).  Other stops on the days were la Playa de Barconaleta (Barconaleta Beach) and Port Olimpic, a beautiful area on the waterfront of what was the Olympic Village of 1992.  There was a festival taking place there celebrating Castilla and Leon, two small villages north of Barcelona near the Spanish border with France.  There were native dancing performances and displays of local customs which our pictures reflect.

   

Tonight I am starting on a backlog of laundry that has accumulated since before I left the United States.  The good thing about this flat is that there is a washing machine on our balcony which prevents us from having to spend time in a laundry mat.  The bad thing is that there is no dryer.  Apparently the custom for this neighborhood is to dry clothes the old fashioned way, via clothesline, so every balcony is outfitted with clothes lines that hang off of the balconies.  The part about this that makes me nervous is that my wet clothes are hanging off of a balcony about five stories up…this may be one way to guarantee that my luggage on the return trip is not overweight.

Keywords: Barcelona, Day 3, Spain

Posted by Marlon Bright | 0 comment(s)

I guess today was Antoni Gaudi day for the four of us as we visited three of his most famous projects.  And let me say that they deserved the hype!  We covered Gaudi in our language and culture seminars and we watched movies and saw pictures of his work, but I have to admit I did not fully appreciate their masterfulness.  To me they were pictures that looked like they were out of a Dr. Suess book and were obviously created by a madman.  I stand corrected this evening as they may have been created by a madman, but that madman was brilliant.  Guadi believed that architecture should reflect nature, therefore, in all of his creations you will not find many right straight cut right angles.  He puts a natural curve to everything which is very creative to have done.  His style was so distinct and unique that it demands to be appreciated.

 We started the day at La Sagrada de Familia which was, and still is in the process of being, Guadi’s masterpiece.  La Sagrada de Familia is an enormous and elaborately designed and detailed cathedral that has a unique place in the Barcelona skyline.  Guadi began construction in 1882, and the construction continues today, some 82 years after his death in 1929.  That’s right, the building is still an active construction site and is projected to be that way until 2029, roughly 127 years later.  Of course there were several delays in construction due to funding and the Spanish Civil War.  The size and detail of this building is like none other.  There are so many different formations on the exterior walls, from knights to lizards to representations of Christ, that it is mind boggling how a person could imagine and execute such detail.  Even the inner supports of the building are not typical the columns twist and come together at the ceiling giving the impression of trees in a forest.  I am sure that I will be talking about such a magnificent building for a long time to come, and I certainly want to come back and see it when it is finally finished.

 

From Guadi’s most famous work we moved to one of his completed works, La Pedrera.  La Pedrera is an actual residence in which people live and work, however, the top three levels are open for exhibition containing a tribute to its architect. The lower of these levels is an exhibition of an apartment in Guadi’s time.  The middle of these levels is the attic of the building complete with all the nature reflected designs of Guadi.  I could go on and on about his mind boggling style, but one gets the picture after a while.  The attic, also, served as sort of the museum of Guadi depicting all of his works.  Guadi was a devout Catholic (hence the reason his masterpiece is a Cathedral) and was very dedicated to uplifting the poorer classes.  In fact, La Sagrada de Familia was part of a plan for a whole neighborhood that included housing, parks, a hospital, and a school.   Another interesting note that I discovered was that our flat was right down the street (literally 4 buildings) from Guadi’s first major work, Casa Vicens.  By far the most interesting thing about La Pedrera was its unique windows and its roof top that was straight out of the Dr. Suess book.  I definitely advise all to check out the pictures.

  

The final stop for the afternoon, La Parc Guell, was nothing short of breath taking.  Again, we had covered this park in our classes, but we really had no idea of the true magnificence of the park.  La Parc Guell sits high above most of the city on a mountain (because it was much more than a hill.  In fact as we exited the metro, there was a series of no less than five escalators and another serious street climb just to get to the entrance of the of the foot of the park.  From there we climbed the trails to its highest point that provided a breath taking panoramic view of Barcelona and the surrounding Catalunya.  We were even able to make some friends along the way that helped us out by snapping a picture for us so that we could all be in one.  At the top we were able to confirm the unique geography of Barcelona.  Barcelona is surrounded by mountains on three sides and the Mediterranean Sea on the fourth.  In fact, the city, itself, climbs up a few of the mountains.  We made our way down from the highest point in the park to its true highlight where Guadi had designed gathering areas and tunnels whose design and detail were rivaled by little other that I have seen.  The views of the city and the waterfront were still insane from here.

    

 

After some serious hiking today, we walked from the park to our flat which we discovered was not all that long of a way.  That’s good because I am sure that I will revisit Guadi’s work and I am sure that I will continue to talk about this day’s experiences in future blog entries.  For now we are preparing to do what we came here to do, and that is seize this great opportunity to learn at the Barcelona Supercomputing Center.  Personally, I know that I have brought my thinking cap and I plan to make the most out of this experience, both at work and at play.  For me being an undergraduate and involved on campus, it is very hard to get too many of the interesting research areas that are being worked on at my university.  This experience will afford me the opportunity to dive in head first and really focus increasing my knowledge and skill set.  It does seem a little daunting, but I and who I have been blessed to be and I know that because of this blessing the possibilities are endless.  This is a great opportunity to see how technology is applied in other cultures and I feel it will be a priceless experience.

I anticipate that I will not be blogging as often (certainly note daily) as things will be getting more exciting and even more busy.  But, for the sake of my recollection and for the benefit of others to come I will continue to frequently record my experiences.

One more note about Barcelona.  Apparently, due to its geography, the weather seems to change sporadically.  It goes from hot to cold to rainy to sunny…all before lunchtime or as the locals call it, siesta.

Keywords: Barcelona, Day 4, Spain

Posted by Marlon Bright | 0 comment(s)

May 20, 2008

I am currently having some technical difficulties with uploading some of my pictures, but here are the links to the albums for the time being:

http://www.facebook.com/album.php?aid=2128823&l=093fc&i

and

http://www.facebook.com/album.php?aid=2128842&l=30e30&i

Posted by Marlon Bright | 0 comment(s)

May 25, 2008

We have been at work now four days.  I feel we are making progress in that we are already interacting with the developers of Paraver and Dimemas to ensure that we can generate our on application performance trace files from our compute cluster in Miami.  I am really enjoying the opportunity to solely focus on research and improving my technical knowledge without being bogged down without having to juggle all sorts of other subject matter.  Today was, also, a great day because Javier and I finally got a key to the room that we are working in.  Prior to this we were at the mercy of the schedule of the other guys working in the room.  We had to wait for them to open the room in the morning and then we had to leave when they left in the evening so that they could lock the door for the night, which meant that we would often have to wait outside for 20 minutes or so when we returned from lunch or “la Siesta”. 

 

I am finding shopping in Spain to be a bit of an adventure.  Besides the fact that the packages are written in a different language, there is somewhat of a difference of the content of the store.  First of all, at least a quarter of every store is committed to wine and beer.  I guess the Spanish enjoy their alcohol.  Then there is the aisle dedicated to oil and vinegar another local favorite.  One other thing they have a lot of is pork.  There are numerous different forms of pork; in fact the local cuisine seems to be “jamon y queso” which to me is just a fancy name for a ham and cheese sandwich.  You can get jamon y queso on everything from a bagel to a pizza.  In addition, I love breakfast and the Spanish don’t put that much of an emphasis as there is little of what I call breakfast food.  Some things that I am missing over here: Cheerios, Quaker Oats, pancakes, corn bread, Catalina French Dressing, etc. etc.  But on the other hand it is good to get used to eating the local favorites such as the hot and fresh bread that comes out of every grocery store on nearly every corner.  Still, however, I count myself as lucky compared to what my counterparts in India and China must be experiencing right now. I am intently proud of myself for being able to successfully send some postcards to America without the assistance of the Spanish speakers, Juan and Javier.  My Spanish is making some progress, but I am still seeking a native friend with the time and patience to help me out a couple of days a week.  This would be a good souvenir to bring home and would definitely come in handy.  Regardless, I am making more of an effort to interact with the locals, something that I was a bit hesitant to do because of my deficiencies in the language. 

This evening I made my first truly home cooked meal in Spain.  It consisted of chicken breast, green beans, and corn.  I did my best impression of Emeril (from the Food network), and I good tell that my suitemates were impressed. After dinner, I went out for a stroll which seems as if it is a common past time for many people in Barcelona.  This time of year, the sun doesn’t set until nearly 9:30pm and therefore you see numerous people out and about until late in the evening.  These walks afford me the chance to really observe the Spanish people and take in a bit of their culture.  Also, today we had internet access installed into our flat which should make communications more convenient.

  Also, today I had a burst in productivity to look into visiting some other destinations in Europe later on in my stay.  So stay tuned for some good experiences and pictures.   One other thing I noted today:  Most of the songs that I hear on the radio or while in the gym are in English.  I would say nearly 40% or more.  This is odd in that from my limited experience, the majority of the population in Barcelona does not speak English.  I wonder if they understand the songs, or what exactly are their thoughts about this fact.

Keywords: Barcelona, Day 7, Spain

Posted by Marlon Bright | 0 comment(s)

May 26, 2008

This was our 2nd action packed weekend in Barcelona.  Friday night, I decided to hit the town by visiting a night spot not too far from our house.  Boy, did I learn some things about the Spanish culture.  First of all, the day is in segments and everything is later in comparison to the typical American Day.  First of all, since we have been here, it seems that the work day starts about 9 or 10 am depending on what your doing (For me I like to get a good start by about 8:45 or so).  Then there is the midday “siesta” or lunch time anywhere between 1:30 and 3:30 which last an hour (or longer for some folks) and then the work continues till anywhere between 6 and 7. The stores stay open pretty late (till about 9 or 10).  Even the neighborhood post office stays open until 8:30pm and the main one until 9:30.  This time of year the sun doesn’t go down until nearly 9:30 which is almost dinner time.  The night spots don’t usually open till about midnight and don’t fill up until about 2am.  The hangouts generally carry over until about 6am. For me, I am not much of a night person, so to experience the culture I just broke the day up into segments.  I got home from the University about 7:15 Friday.  I then took a nap from about 7:45pm to 9:15 or so.  I woke up, fixed some dinner, showered, and did some searching on the computer.  I headed out again at about 11:30 to discover that I was so very early to most of the night spots.  I managed to make two friends who were from Barcelona and we hung out for the night.  It was a lot of fun getting to know the locals even though there was a significant language barrier.  Most people got out to where they were going about 1:30am or so.  Me, I was on my way home by 2:30 or so to get ready for the next day.  I wasn’t really tired that night and I woke up very refreshed and ready to roll at about 10.  I have found the weekend formula.  Just break up your sleeping into two parts! 

We spent most of Saturday at a high altitude.  We took a cable car from La Playa de Barconaleta over the Port of Barcelona to midway up the majestic Montjuic which overlooks the city and the Port.  From there we hiked a bit up to another cable car to take us to the top of the mountain to La Castell de Montjuic, a commanding military fortress at the very top of the mountain that commanded a dominating presence over the city and the port.  What a site this was to see completely outfitted with heavy artillery and some seriously fortified walls.  Any opposing military trying to take Barcelona from the sea would have a very hard time getting through the blow this installation could bring!

  

Heading down the mountain on the other side was the impressive Olympic complex from the 1992 Summer Olympic games complete with la Estadi de Olympic (Olympic Stadium) as well as various other sporting venues for sports such as basketball, baseball, swimming, gymnastics, etc…  One could not help but be in astonishment at this complex built into the side of a mountain with such sprawling and magnificent scenary.  From there we worked are way down to Placa de Espanya taking in the sites of the National Art Museum of Catalunya and the Font de Montjuic (a huge fountain).  However, we have to return there in the evening as we have been told of its picturesque.  We then went to check out a soccer game taking place at Camp Nou (Barcelona’s main futbol stadium seating over 120,000 people) where a match between Catalunya and Argentina was taking place.  However, we decided not to go in because the cheapest ticket was €30, money we felt we could spend when the main team, FC Barcelona, was playing.  We then headed back for our naps, and I headed out again this time for Las Ramblas.  There wasn’t much going for me this evening so I decided to call it an early night (in Spanish standards).

       

It rained most of Sunday, but Javier and I did go to the Maritime Museum of Barcelona which was quite nice. We, also, had planned on going to the big bullfight that night, but it was canceled, we suspect due to the weather.  There were some good things that did come out of Sunday.  I booked a 3 night trip to Rome in June and a 2 night trip to Lisbon, Portugal in July.  So, I look forward to some more great fun on this side of the pond (Atlantic Ocean).  And we will certainly heed to the common saying, “While in Rome, do as the Romans do!”

And I have started another photo album, so to check the album out in full go to:

http://www.facebook.com/photo.php?pid=34042699&l=cd05b&id=18715815

Keywords: Barcelona, Day 11, Spain

Posted by Marlon Bright | 0 comment(s)

May 27, 2008

ACTIVITIES:
In this part, please mention your activities during the last week.  This past week we worked with the guys in our office to get mpidtrace working properly on the GCB and Mind clusters in Miami in order that the suite of application profiling tools (Paraver and Dimemas) would be working properly. We, also, went over some preparation pointers for running Amon and Aprof on MareNostrum.   I, personally, am in the progress of further and in-depth study of Linux operation and other related project related technologies in order to be better able to operate in the Linux environment.  In addition I reviewed three papers. We, also, took the time to map out or goals (shown in plans section) for our stay in Barcelona in order to mark our progress.  We have found the guys in our office very helpful in working towards the first two.
 

ACCOMPLISHMENTS:
In this part, please only mention your significant accomplishments that you have not reported previously. Note that you do not have to have an accomplishment each week. 
  • Access to network
  • Account creation in MareNostrum


ISSUES/PROBLEMS:
In this part, please mention the minor and major problems that you are facing in your research.  

PLANS:
In this part, please mention your short- and long-term plans for the weeks and months to com.  Barcelona Research Objectives 

Overall Objective: Achieve accurate and timely performance prediction on compute cluster to be used in Grid Computing Environment Meta-Scheduling. 

1.      Get mpidtrace linking properly with WRF compiled on GCB, then Mind.

2.      Use generated MPI tracefiles (Paraver and Dimemas) to do prediction between Mind and GCB

3.      a) Install Amon and Aprof on MareNostrum.

b) Run benchmarks on MareNostrum

4.      Experiment with how well Amon and Aprof scale to larger number of nodes

5.      Analyze how Amon and Aprof relate to/could possibly combined with Dimemas.

6.      Work with  Marc to see how we can optimize the gridification of WRF 

SUMMARIES/CRITIQUES OF PAPERS: 

In this part, please include a short review of the papers you have read during the last week. The review should include three short paragraphs for each paper. The first paragraph should be a short summary of the paper; The second paragraph should include a short critique of the paper; and The last paragraph should include a discussion on how this paper is related to your research. 

Review: The Vision of Autonomic Computing 

The essential idea of autonomic computing is that systems are able to manage themselves according to a system administrator’s policies and goals.   This should free the system administrator up from many time consuming tasks and should ensure peak and optimal performance for the system at all times by reducing opportunities for error. The four concepts of self-management are self-configuration, self-optimization, self-healing, and self-protection.  The term autonomic sums up the direction in which efforts are being made.  They are being made to take most of the human element out of the configuration of components and systems, performance improvement and optimization, error detection and repair, and security.  Furthermore, things are complicated when taking into consideration the interaction of separate heterogeneous autonomic systems which may be located across company and country boundaries. 

This article described the vision and need for autonomic computing.  It brings up the important fact that we are now reaching the limits of our computational limits as we know them and that autonomic computing is our only option left.  All this text did not explicitly cover this, it did say that grid and web services fell underneath this umbrella.  Autonomic computing seems to me to be a step in the direction towards artificial intelligence.  With this said, the article makes sure to mention at the end the people that will need to be involved in the development of this process.  These people were not only computer scientists and engineers, but they were scientists, psychologists, economists, and people in the legal profession.  Many different factors have to be taken into account such as robustness, costs, security, effect on humans, and legal aspects.  With this advancements have to be made carefully in this direction to assure that humans to not develop things too big for them to handle. 

This paper lays the basis and overarching foundation for my research.  Moving down the hierarchy, my research then falls under grid computing then to job-management then to meta-scheduling then to application cluster performance prediction.  These are all essential topics in a move towards autonomic computing.  Our goal is to take job scheduling to the grid environment and develop applications that help meta-schedulers make better scheduling decisions to optimize the performance of applications across the grid.  From there it will be the individual systems’ responsibility to autonomically manage the jobs, its performance, and its security.   If there is a problem with the system, then the scheduler should recognize this and not submit any jobs to the system until the issue is resolved. 

Reference: 

Jeffrey O. Kephart and David M. Chess. The vision of autonomic computing. IEEE Computer, 36(1):41-50, January 2003.

Review: Transparent Grid Enablement of Weather Research and Forecasting          

This paper gives a high level overview of the current status of the transparent grid enablement of the Weather Research and Forecasting code.  It goes into great detail in describing the motivation for this project as being the accurate and timely prediction of such wether disaster phenomena such as hurricanes.  It details the need for specific zip code level forecasts for the benefit of meteorologists, business owners, and emergency response personnel. From there it talks about the different aspects of the project such as the transparent grid enablement through such technologies as TRAP/J and Grid Superscalar.  Other aspects of the project are then described such as the middleware, meta-scheduling, job flow management and profiling tools. 

This paper serves to lay out the over arching foundation and motivation for my current research.  My current research is in the area of profiling, optimizing the tools Amon and Aprof in considering more compute characteristics and offering a better comparison with the prediction tool Dimemas. 

Reference: 

Transparent Grid Enablement of Weather Research and Forecasting S. Masoud Sadjadi1, Liana Fong6, Rosa M. Badia2, Javier Figueroa1,9, Javier Delgado1, Xabriel J. Collazo-Mojica8, Khalid Saleem1, Raju Rangaswami1, Shu Shimizu4, Hector A. Duran Limon5, Pat Welsh3, Sandeep Pattnaik10, Anthony Praino6, David Villegas1, Selim Kalayci1, Gargi Dasgupta7, Onyeka Ezenwoye1, Juan Carlos Martinez1, Ivan Rodero2, Shuyi Chen9, Javier Muñoz1, Diego Lopez1, Julita Corbalan2, Hugh Willoughby1, Michael McFail1, Christine Lisetti1, and Malek Adjouadi1 1: Florida International University (FIU), Miami, Florida, USA; 2: Barcelona Supercomputing Center, Barcelona, Spain; 3: University of North Florida, Jacksonville, Florida, USA; 4: IBM Tokyo Research Laboratory, Tokyo, Japan; 5: University of Guadalajara, CUCEA, Mexico; 6: IBM T. J. Watson, NY, USA; 7: IBM IRL, India; 8: University of Puerto Rico, Mayaguez Campus, Puerto Rico; 9: University of Miami, Coral Gables, Florida, USA; 10: Florida State University, Tallahassee, Florida, USA  

Review: A Modeling Approach for Estimating Execution Time of Long-Running Scientific Applications

 This paper describes a modeling approach for the estimation of application execution times on a compute cluster contained in the software application tools Amon and Aprof.  While there are several factors that should be taken into account when considering the run time of an application.  A first assumption was made that the most significant contributors are the number of processors and the CPU clock speed.  To summarize the details, the following model was used:in which T­exec is the execution time, Cis the i-th contribution, and m is the number of the contribution terms.  Essentially, in this case, there was C0 and C1 due to the degree of parallelism (number of nodes) and the CPU performance contribution.  For this paper the final model expanded to being: where α0, α1, β0, β1, we constants such as execution overhead and application characteristics. 

Experiments were performed on two clusters at Florida International University, GCB and Mind, with 8 nodes and 16 nodes respectively.  Experiments were executed varying the number of nods and the CPU utilization.  Monitoring information from Amon was used as input data to Aprof and prediction were made for within the same cluster and intra-cluster.  All predictions were made in less than 10% error of their actual values, thereby supporting the model. 

The Amon and Aprof approach differs from other related works.  For instance, it differs from the performance prediction simulator, Dimemas [2], in that it focuses on online prediction.  Yang et al. [3] requires the analysis of the application source code and a sample execution on the target platform. The Amon and Aprof approach differs in being application agnostic, doesn’t require a sample execution on the target platform, and models execution scale addressing distributed applications.  An interesting point here is the fact that the experiments were performed on small clusters.  We now want to make benchmarks on a larger number of resources of which it will now be beneficial to access the supercomputer MareNostrum.  However, these results will be used to optimize the model that will be used in grid environments. 

My current research is an extension of this work.  What we want to now do is validate this model for a larger number of resources, and get a more in-depth comparison with the performance simulator Dimemas.  We are, also, are considering adding in memory parameters into Aprof’s prediction model as part of the optimization process that is discussed.  In addition, this research focuses on the Weather Research and Forecasting Code of which it is our ultimate goal to grid enable this application.  Down the road, however, we would like to further validate our assumption that this model should be application agnostic and further its application. 

References: 

[1] S. Masoud Sadjadi, Shu Shimizu, Javier Figueroa, Raju Rangaswami, Javier Delgado, Hector Duran, and Xabriel Collazo. A modeling approach for estimating execution time of long-running scientific applications. In Proceedings of the 22nd IEEE International Parallel & Distributed Processing Symposium (IPDPS-2008), the Fifth High-Performance Grid Computing Workshop (HPGC-2008), Miami, Florida, April 2008. 

[2] R. Badia, F. Escale, E. Gabriel , J. Gimenez, R. Keller, J. Labarta, M. S. Müller, Performance Prediction in a Grid Environment. European Across Grids Conference, 2003. [3] L. T. Yang, X. Ma, and F. Mueller. Cross-Platform Performance Prediction of Parallel Applications Using Partial Execution. Proceedings of Supercomputing, 2005. 

Keywords: 23 May 2008, Barcelona, PIRE, Spain, Weekly Status Report

Posted by Marlon Bright | 0 comment(s)