New domain!
This blog was moved to http://www.marioalmeida.eu/ !
Offline Routing and Regenerator Placement and Dimensioning for Translucent OBS Networks
Been a long time 😉 Seems I have been having a few days of rest (or the closest possible to it since I only managed to go to beach once!!).
Anyways, I’ve been doing a project that consists on implementing and evaluating the ILP and heuristics presented on paper “Offline Routing and Regenerator Placement and Dimensioning for Translucent OBS Networks“.
So what is a Translucent network?
…both the electronic switching networks and all-optical transparent networks have their own pros and cons. A translucent network just compromise these two types of networks to take the advantages of the two types of networks to achieve a good balance between cost and network performance. For cost saving, the translucent networks maximally pursue optical bypass opportunities when establishing lightpaths… (read more : here)
Is it useful?
The deployment of translucent optical networks is considered the most promising short term solution to decrease costs and energy consumption in optical backbone networks… (from : paper)
What is its contribution?
…In the T-OBS network the problem of routing and regenerator placement and dimensioning (RRPD) emerges.
…RRPD is a complex problem and, in order to approach it, we propose to decompose it into the routing and RPD subproblems. As a consequence, we provide a mixed integer linear programming formulation of the routing problem and several heuristic strategies for the RPD problem.
…we show that the proposed RPD heuristics ensure a proper quality of transmission performance whilst at the same time providing a cost-effective T-OBS network architecture. (from : paper)
The paper presents different heuristics and respective evaluations. For the heuristics we’ve decided to implement the Biased random-key genetic algorithm (BRKGA) since its the most efficient and the one that generates the best results (generally optimal). I’ve found a very informative presentation on slideshare:
High Availability of Services in Wide-Area Shared Computing Networks
I finally managed to put together the final report for the research that I have been doing. Unfortunately, we only managed to get access to planetLab less than a week before the final delivery (sending lots of emails to multiple entities). We could only do minor experiments on it. As always, I am a bit exigent and I feel that I could have accomplished much more with it. Still I think it’s an entertaining paper.
Here is the abstract:
Highly available distributed systems have been widely used and have proven to be resistant to a wide range of faults. Although these kind of services are easy to access, they require an investment that developers might not always be willing to make. We present an overview of Wide-Area shared computing networks as well as methods to provide high availability of services in such networks. We also make some references to highly available systems that are being used and studied at the moment this paper was written.
And here is the paper:
Performance evaluation of parallel raytrace execution
I can finally post the presentation on the small project I’ve been working on for the last week. I will be producing a more detailed report on it in a near future.
Always look through the bright side of life
I felt like writing my first less formal post of this blog to talk a bit about life and how it feels good when you’ve got a filled up Agenda.
These weeks have had their ups and downs like everything else. I was rejected for google soc. Although i feel sad for it, it also makes me feel relief because I can still work on it but with more time to put more effort into the projects I have at the moment. I am preparing to start a new project for a video game with my flatmate, I had a nice idea for a gameplay that i feel that could be innovative and interesting and it is in a area that my colleague wanted to work on and has some experience in. I don’t know if it will work or if it will be addictive enough but that’s why prototyping exists! I’m exited with the prospects!
Three projects are still undergoing: parallel raytrace instrumentation and measurements; high availability in Wide Area Networks and a still to define better ILP and heuristic/methaheuristic project. The first will finish soon and I hope I can post the results by tomorrow.
Also I have been planning some trips to relax and enjoy, namely to Tenerife/Gran Canarias and to Turkey. Either to go to the beach or meet some friends that I met during my European master journey.
I leave you with a song that I love to sing to myself whenever something bad (sometimes good) happens!
Cloud scale identity fabric

Architecting a cloud scale identity fabric
WiiKill Asteroids
Just came across with my first game ever! I done it in 2008! I had a terrible methodology of keeping track of my projects at that time (SVN was alien technology i guess :P). I had lots of fun doing and playing the game and hope you guys enjoy and take into consideration that this was my first contact with OpenGl (C++).
There are 3 different control modes:
- Default: use the arrows to move the spaceship and space to shoot! (see readme for more details – pt)
- Partial mouse mode: press F3. Now spaceship points to mouse and you can click to shoot!
- Full mouse mode: press F3 then F4: Now the spaceship also follows the mouse. Its acceleration depends on the distance to the mouse (somewhat like an elastic)!
Credits also go to two of the greatest persons I ever worked with, José Lourenço and Tiago Freitas.
Busy project agenda
I haven’t been very active in the blog lately since I had to interrupt the decentralized systems project on high availability.
I have another project in between that consists on using parsec benchmark suite on a multiprocessor system and analyze its performance using paraver. Just got parsec running after a few tweaks and managed to change its build configuration to include the extrae libs so I can soon see some nice graphs on paraver.
Boada system is composed by 2 processors of 6 cores each running @2.4GHz. It uses hyperthreading so the total of cores is 24. Would be cool to use it for something else 😛
On another topic, I have applied for google summer code. I saw some really nice projects there and I wanted to apply to all of them.. but its always better to choose one and focus on the proposal. I am sad that I didn’t have time to research before applying (even more in the case of Tor project). But I did apply for a very cool project called Battle for Wesnoth, I am a fan of the game. I don’t think I’ll be selected though because this week I don’t have enough time to be more in contact with the development team and hack some fixes for the game. Maybe I’ll keep connected with project even without the google incentives.
Cheers to all!
Decentralized Systems Project on High Availability of Services in WAN
As it has been noticeable in the recent posts I have been performing some research on high availability of services in Wide-Area Shared Computing.
I have been writing a report on the topics I’ve researched so far and I intend on writing a big post in this blog about it. If you are interested in the results of my research so far you can follow its development on the following document: report.
So far it describes Wide-Area Networks as well high availability mechanisms such as migration, replication, quorums consistency and group memberships. It will have references to systems like Dynamo from Amazon and Zookeeper from Yahoo.
Keep connected 😉
Autonomically Scalable Services in WAN
So I’ve got only three days left to research on the area and fortunately a big headache as well. Nothing like a pill, coffee and we’re good to go! At least I can blame any inconsistency in what I write with my physical state 😛
After reading “Resource-Aware Migratory Services in Wide-Area Shared Computing Environments”, I took a look at the paper I mentioned previously : “Building Autonomically Scalable Services on Wide-Area Shared Computing Platforms”. The papers are from the same authors and basically covers the issues I mentioned in the previous post, such as the lack of proper policies and metrics.
The authors claim this paper contribution is:
- Models for estimating service capacity likely to be provided by a replica in the near future.
- Models for dynamic control of the degree of service replication in order to provision the required aggregate service capacity based on the estimated service capacities of the replicas.
- A reliable registry service for clients to locate service replicas.
The results show that this system can predict capacity with 95% accuracy, which doesn’t seem bad, depending on the threshold at which the service is replicated.
Overall there were some topics that i found interesting:
- The load balancing is done at two levels: registry level and node level. At the node level this load balancing can be achieved by either redirecting requests or redirecting clients. Also, the replicas send tokens to each others in order to estimate how much requests they can redirect to each replica.
- They kept the deployment agents, this agents keep a soft state of the information of replicas and are responsible for the dynamic control of the degree of replication. They create, launch and detect failure of replicas. Interestingly, they create a recovery agent that is responsible for monitoring the deployment agent and revive it if needed. If shutdowns in PlanetLab or in WAN are somewhat frequent, then the probability of both going down in a short period of time shouldn’t be that small. This could render the service vulnerable.
- They seem to be able to read my mind (before I even think about it), and so they decided to allow reads on the replicas in order to have a better load balancing. The way this replication is done is still not very explicit, maybe I have to go back to the paper that describes Ajanta.
I know I had a few more things to say about it but today I’m the one with limited resources, both mental and physical.

