Distributed Algorithms - Science topic

Hi Arijit Nandi ,

Thanks for your input. I have their codes. They have followed the same. I have attached their code below.

I'm not keen on calling anything "smart". Any method will fail under some circumstances, such as for some outlier that no-one have thought of.

Himadri Nath Saha, it seems to be a fascinating motivation scenario. The investments and the continuous growth of streaming games have imposed new requirements of response time (i.e., ultra-low latency). I am looking for DSP systems where latency-sensitive applications can be easily deployed on geo-distributed constrained resources avoiding creating a multi-tier of components.

Amal

There must be an improvement since with MAS you have means to negociate, cooperate, that do not exist whithin ant colony alone. You must ask yourself on the existance of works that enforce ant colonies with MAS. The conclusion is that MAS are generally used to facilitate Ant colony communication and automation.

Hope this will help you

Hakima

Thanks a lot Sir. Please refer me if you can.

@Zeashan Khan

Thanks a lot.

This paper might help you..

Simulation of a Video Surveillance Network Using Remote Intelligent Security Cameras

Multi-agent systems run on the distributed environment (Sometimes in a single computer) capable to communicate with each other through the message parsing. Through the communication, agents can solve complex problems easily. If you use a correct framework it is not probably slower. However, algorithm processes are strata forward. Note that MAS has the different approach than the algorithmic process. In addition, MAS can provide results that cannot previously predict. Try it.

Dear Nicky,

I don't know much about Fuzzy Cognitive Mapping or Normal Distribution Algorithm, so I can only give you some hints about Factor Analysis, but I would say that first of all the method you choose should fit your research question and not only the type of data you have.

If you aim to looking for latent structures and structural factors, you could use Factor Analysis. FA works with ordinal data if you base the construction of factors on polychoric correlations. The lavaan package in r allows you to do so, if you specify you have ordinal variables: http://lavaan.ugent.be/tutorial/tutorial.pdf

Good luck!

Mireia

Check here:

You can get some  help from above link. 

This is a good reference for your problem..

Your data is partitioned on both on time and id. so you can say 

You wanted to calculate the cross correlation matrix. node A({id1, Jan}{id2,Jan}), Node B({id3, Jan}, {id4,jan}) again for feb node A({id3, feb}{id2, feb}), node B({id1, Feb}{id2,Feb})

Jan,Feb : Data generated in months on jan and feb. You can say id1, id2 are data from different sensors.

you can check this:

Tauer, G., & Nagi, R. (2013). A map-reduce lagrangian heuristic for multidimensional assignment problems with decomposable costs. Parallel Computing, 39(11), 653-668.

Chicago

In summary: for very very large p2p systems (file-sharing), I would go with a reputation system / shared history; for small ones, I would go with accounting, bilateral interactions (private history).

Commonly we can find different choises in the application of Approximation algorithm. Three most used are:

When we inspect the theoretical performances, we find out that the best one is the first algorithm.it has been observed the worst performance with the BFS tree based algorithm. The reason is that asynchronous BFS construction is a costly operation.

I hope it will be helpful.

best regards

Hi Dimitry,

There exist many simuators and simulation frameworks (omnet++, NS2 , NS3, simgrid ...). So it is difficult to give a general answer. It depends on the pupose of simulation :

With all those questions, there is an other one, are you looking for a simulator in a special language ?

If the C++ does not mind, you can use omnet++ ( https://omnetpp.org/ ) . This simulator can be used for algorithms simulation, or protocols simulations, offer a nice programming environment, and allow visualization of the executions.

Best regards

Olivier

It is right, thank you a lot.

Consider “making a simulation of a petri net” to be the same as “writing a computer program using petri nets”. To use petri nets for writing a computer program is to organize a computer program using Petri Net elements and annotations. To create this organization, first establish the relations of computations in terms of places, transitions, inputs and outputs. Second create annotations of places, transitions, inputs and outputs for the computations.

Consider making a simulation of a countdown timer. The countdown timer begins from a specified value (say, 10). Using some interval, the timer value decreases by one. When the timer value reaches zero, the countdown ends.

This computation may be modeled in terms of a place, an input and a transition. The place has a mark annotation, a value that represents the timer value – the place is the timer. The input has a “fire” annotation, a computation that deducts one from the input place mark annotation. The input also has a status annotation, a ‘true’ value means the computation may proceed and a ‘false’ value means that the computation should not proceed. The input has an “is enabled” annotation, a computation that determines the value of the status annotation – in this case, if the mark annotation of the input place is greater than zero, the status annotation of the input is ‘true’ and ‘false’ otherwise. The transition has “fire” annotation, a computation that delegates the “fire” computation by “firing” its input. The transition has an annotation status, a true value means that the value of its input status annotation is ‘true’ and ‘false’ otherwise. Consider too visual annotations of the net elements: a circle for the place, a square for the transition and an arrow from the circle to the square for the transition, and dots for the mark annotations of places – empty for 0 mark, one black dot for 1 mark, two black dots for 2 marks, etc. [The types of annotations considered so far are those found on Place/Transition nets, a kind of Petri Net.]

To make the “simulation” interactive, consider an event annotation for the transition, a computation that delegates a user event (such as a mouse up event) to “fire” the transition. Furthermore, consider a visual annotation for the status of the transition – green if the status annotation is true and empty otherwise.

Based on this example, I created an interactive “simulation” of this system in PDF. Graphics were created using PowerPoint and exported as PDF. Computation logic was implemented as JavaScript programs. And interactions were integrated with the computation logic using the Acrobat/JavaScript API. [See link attached to this reply].

For petri nets with many net elements and annotations, several other issues (such as naming conventions, high-level graphics conventions, and software engineering workflow processes) must be addressed. Here is a research paper related to this topic – Net Elements and Annotations: Computations and Interactions in PDF. [See link attached to this reply].

i am taking bus data and line data of 118 bus system and enter to my load flow and run the load flow getting base case power loss is 1291kw but compared some papers it is 1296kw which is correct?

You clan try lemon or igraph even if all the algorithms are ont parallelized. However, y ou clan take a look to bob++ (google : Bertrand Lecun, bob++) .

@Peter: I think that the way this question is asked is not so much about consistency in the DB. Though consistency is certainly a problem with this approach.

I can't think of any situation where this kind of multi-writer approach is useful. Suppose the current state is A and writer1 wants to write B, writer2 wants to write C. In many cases writers think that the current state is what they have written. Sometimes this cannot be assumed, so each writer reads back the current state directly after writing. Suppose that writer1 wins the battle and writer2 will never write C (i.e. it is hidden, just as you explained). Now writer2 wants to read back the current state, but writer1 does not have finished writing. Which state does writer2 get? A, B or C? (B only works if writer2 waits until writer1 is finished -> you need synchronization again.) Since C will be hidden it only can be A or B. We can't read A because writer2 knows that it is older than C. And B is not available yet. What do you do now?

In another scenario, maybe when writing to a log file where there is no reader during execution of the application, I don't see why we would throw away any writes. We could use one thread that handles a queue and both writers can enqueue their write commands almost without any delay (lock-free data structures come to mind).

Dear Kareem

Salam,

According to above question, i have a auto-scaling system with swinging input volume and i want to allocate my resource in the best way.

On the other hand the pattern of my input volume is unpredictable however, is there any algorithm that can help?

.

minhash is indeed a locality sensitive hashing scheme when locality is defined according to the jaccard similarity

see (by increasing completeness) :

http://en.wikipedia.org/wiki/MinHash#Jaccard_similarity_and_minimum_hash_values

http://www.cs.columbia.edu/~coms699812/lecture1.pdf

http://infolab.stanford.edu/~ullman/mmds/ch3.pdf

and

http://www.ucd.ie/t4cms/CeADAR_RS_Boydell.pdf

for an example of implementation

.

I'll try to answer in layman's terms (no guarantee of being formally correct).

Distribution has to do with where the computation physically resides. A distributed algorithm is executed on multiple CPUs, connected by networks, buses or any other data communication channel. Parallelism has to do with the fact that in the algorithm two or more flows of control may execute (even if only virtually) at the same time.

True (not virtual) parallelism requires distribution. But parallel algorithm may execute also on virtual machines realized on a single physical CPU.

platform as a service from open source community and redhat: openshift

this p-a-a-s software is included in linux distribution fedora 19.

You need a correct mathematical model of the DG operation problem, including the objective function(s) and system econ-tech constraints and acceptable modeling of uncertain parameters, then you should either code a numerical optimization algorithm or use a commercial package (like GAMS) to solve the hour-based daily DG scheduling problem. Then analyze the results to see if they make sense. If they did not, update your model/math prog until you get meaningful results for the next 24-hours. Good luck!