Today marks the official start of Summer of Code 2008, and I'm looking forward to spending more serious time on the project from now on. I still have my exams on the horizon (the last one on June 5:th) as well as some other things to take care of before the end of the semester, but things should be winding down on the school-front from now on.
The main cause of this winding down is the fact that our lab assignment (writing a chat-server + client) got handed in this last Friday. The client-side of that lab can be tried out here (chat servers by other groups can probably be found at the default location over the next couple of weeks, though I take no responsibility for how well they work).
On a tangent, doesn't end-of-semester suck? Or, well, not really. It's actually made up of a bunch of really awesome parts, but the fact that everything conspires to happen at once sucks. At the same time that you have to wrap up all the things you've been doing for the last few months, you want to get off on a good start on whatever you'll be doing over the summer. Also, this is when you really want to be out barbecuing and hanging out with all the people you won't be seeing for the next few months (that's not likely to happen, I'm afraid).
To top this all off, my part time job really wants me to work during all of this, so I'm in for a hectic couple of weeks.
Oh, and I got my start-of-program gift from Google today! Looks interesting...
Monday, May 26, 2008
Monday, May 12, 2008
Update
This past weekend I was away doing some work for the student union.
While I do sometimes do actual work for the student union, this time it largely consisted of laying out in the sun during the day reading documentation about GridFTP and XIO (and in the evening, sauna and beer). Nice weekend indeed.
Last week I also discussed some aspects of the project with my mentor, in particular the possibility to add some intelligence to the compression. One example of this would be to let the driver have access to a number of compression algorithms and do some analysis of the data before determining which one to use. Another, simpler, method would be to let the driver keep track of of the compression ratio of the last few blocks and stop compressing if it's below a certain threshold (i.e., just wasting CPU cycles).
Now, there is a whole bunch of questions that need to be answered before I can say if this is a) a good idea and b) feasible within the time-frame of the project. But it does sound neat, so I will definitely look into this further.
Other than that, my time right now is mainly consumed by course-work, in particular my current lab assignment. This will likely be the case until the end of the semester, but I can probably still sneak in a bit of GSoC time here and there.
While I do sometimes do actual work for the student union, this time it largely consisted of laying out in the sun during the day reading documentation about GridFTP and XIO (and in the evening, sauna and beer). Nice weekend indeed.
Last week I also discussed some aspects of the project with my mentor, in particular the possibility to add some intelligence to the compression. One example of this would be to let the driver have access to a number of compression algorithms and do some analysis of the data before determining which one to use. Another, simpler, method would be to let the driver keep track of of the compression ratio of the last few blocks and stop compressing if it's below a certain threshold (i.e., just wasting CPU cycles).
Now, there is a whole bunch of questions that need to be answered before I can say if this is a) a good idea and b) feasible within the time-frame of the project. But it does sound neat, so I will definitely look into this further.
Other than that, my time right now is mainly consumed by course-work, in particular my current lab assignment. This will likely be the case until the end of the semester, but I can probably still sneak in a bit of GSoC time here and there.
Tuesday, May 6, 2008
First peek at source code
Nothing mayor to report. I still don't have a whole lot of time to spend on the project, but I try to sneak in an hour here and there. I don't think this is an issue as far as the project is concerned, but I regret that I haven't had much time to spend on community bonding yet (I stuck my head into the Globus Pub at the MCS MUD a few days ago, but that's about it).
This past week I've begun peeking at the source code, along with reading more documentation. I was pleasantly surprised to see that there is an implementation of a minimal driver which does nothing but pass data along. That sort of thing is always helpful in understanding how something works. Also, I think it will make a good starting point for developing the compression driver (in fact, throwing together a do-nothing driver was what I had in mind as my first mini-milestone).
This past week I've begun peeking at the source code, along with reading more documentation. I was pleasantly surprised to see that there is an implementation of a minimal driver which does nothing but pass data along. That sort of thing is always helpful in understanding how something works. Also, I think it will make a good starting point for developing the compression driver (in fact, throwing together a do-nothing driver was what I had in mind as my first mini-milestone).
Subscribe to:
Posts (Atom)